AIR POLLUTION AND ENERGY EFFICIENCY. Participants' comments provided in the Correspondence Group on EEDI review beyond phase 2. Submitted by Japan

Transcription

1 E MARINE ENVIRONMENT PROTECTION COMMITTEE 72nd session Agenda item 5 MEPC 72/INF.12 1 February 2018 ENGLISH ONLY AIR POLLUTION AND ENERGY EFFICIENCY Participants' comments provided in the Correspondence Group on EEDI review beyond phase 2 Submitted by Japan SUMMARY Executive summary: Strategic direction, if applicable: This document provides participants' comments provided in the Correspondence Group on EEDI review beyond phase 2 established at MEPC 71 3 Output: 3.6 Action to be taken: Paragraph 3 Related documents: MEPC 70/5/15, MEPC 70/5/16, MEPC 70/5/17, MEPC 70/5/18, MEPC 70/INF.29, MEPC 70/18; MEPC 71/5/2, MEPC 71/5/6, MEPC 71/5/7, MEPC 71/5/12; MEPC 72/INF.7; MEPC 71/INF.16, MEPC 71/17; MEPC 72/5/4 and MEPC 72/5/5 Introduction 1 The Correspondence Group on EEDI review beyond phase 2 was established at MEPC 71 under the coordination of Japan. The progress report is set out in document MEPC 71/5/4 and the coordinator's summary is set out in the annex to document MEPC 71/5/5. This document contains participants' comments provided in the Correspondence Group and updated information on energy saving. 2 The list of annexes is as follows: Annex 1: Comments submitted from participants for round 1; Annex 2: Comments submitted from participants for round 2; Annex 3: Summary of latest information on energy-saving ; I:\MEPC\72\MEPC 72-INF.12.docx

2 Page 2 Annex 4: Annex 5: Annex 6: Annex 7: Annex 8: Annex 9: Annex 10: Annex 11: Annex 12: Annex 13 Annex 14: Latest information on energy-saving ; Table of correspondence between ice classes of different classification societies; Contribution to request 1-1 of the first round by the United States; Contribution to request 2-3 of the first round by the Netherlands; Contribution to requests 4 and 5 of the first round by Russian Federation; Contribution to request 1 of the second round by Japan; Contribution to requests 1, 2-1, 2-2, 2-3, 2-4 of the second round by Norway; Contribution to requests 2-1 and 2-2 of the second round by the United State Contribution to request 2-4 of the second round by ICS; Contribution to request 4 of the second round by the Russian Federation; and Contribution to the latest information on energy saving by Finland Action requested of the Committee 3 The Committee is invited to note the information provided. *** I:\MEPC\72\MEPC 72-INF.12.docx

3 Annex 1, Page 1 ANNEX 1 COMMENTS SUBMITTED FROM PARTICIPANTS FOR ROUND 1 Request 1: Members are requested to provide views/comments and data on the following points. The latest EEDI data contained in the EEDI database is distributed as Annex 1 to this coordinator s remarks. 1-1 : Do you have any concrete suggestion about the analysis of EEDI database for review exercise? If so, please provide your proposal and an example of analysis for consideration of the Group. Submitter Australia China Denmark Comments (with rationale) No comment In the review process, it can t be only seen the total satisfaction percentage of whole ship numbers in every ship type. Because sometimes even the total percentage looks very high, but in some segments the situation is completely different, such as large bulkers in bulk carrier and large tankers in tanker. In addition, it should be clearly understood what kind of methods could be used in the EEDI further reduction and how much improvement could be achieved especially for those traditional method such as hull optimization, energy saving device and propeller design and etc. Because many current new ships have almost no room to get further improvements on the methods they have already chose. The progress regarding attained EEDI is presented graphical in the figures of Annex 1 - EEDI data as of 1st Aug The figures show the attained EEDI compared to the different phases of EEDI reduction. However, referring to figure 1 below, it seems quite difficult to draw clear conclusions based only on this graphical presentation, especially on ships with low values of DWT.

4 Annex 1, Page 2 Submitter Comments (with rationale) Fig 1 Attained EEDI for tankers during phase 0 and phase 1 Another way to present the progress of EEDI would be to show the attained EEDI relative to the EEDI reference value. The relation between attained EEDI and reference EEDI value are also listed in the referred Annex 1. A graphical presentation of Container vessels, Tankers and Bulk Carriers are shown below in figure 2, 3 and 4. Please note, however, that the attained EEDI in the annex

5 Annex 1, Page 3 Submitter Comments (with rationale) is calculated relative to the value of the applicable phase and not to the reference line. This gives false results for the relative phase 1 values. To give the correct presentation, the relation for phase 1 is re-calculated and shown in the figures. Fig 2 EEDI of Container ships - relative to EEDI reference line value

6 Annex 1, Page 4 Submitter Comments (with rationale) Fig 3 EEDI of Tankers - relative to EEDI reference line value

7 Annex 1, Page 5 Submitter Comments (with rationale) Fig 4 EEDI of Bulk carriers - relative to EEDI reference line value Finland We have the following general comments on the use of the EEDI database for the review exercise. The analysis could be done based on the statistical data and/or based on a more detailed analysis of the potential possibilities of certain ship types and sizes to meet phase 3 requirements of EEDI taking into account application of the energy saving applicable to the ship type and size. In our view, the ship types included in the analysis (see Annex 1, EEDI data as of 1st of Aug 2017 ) can be roughly divided into two categories: slow ships, i.e. ships with a maximum service speed of 15 kn or less, and fast ships, i.e. ships with a service speed more than 15 kn. Bulk carriers, tankers and general cargo ships belong to the category of slow ships, and the

8 Annex 1, Page 6 Submitter Comments (with rationale) other ship types, i.e. gas carriers, container ships, refrigerated cargo carriers, ro-ro cargo ships (vehicle carriers) and ro-ro cargo ships belong to the category of fast ships. 1. Slow ships We agree that the EEDI regulations should trigger improvement of energy efficiency of ship, but maintaining the safety of shipping has to be the overriding priority. Therefore, especially with regard to slow ships, it is clear that the new minimum power requirements given for tankers and bulk carriers should be taken into account, when the possibilities of these ship types to meet the phase 3 EEDI regulations is evaluated. We would like to point out that, even if these ship types could meet phase 3 requirements of EEDI, these ship types may not be able to meet the minimum power required for safety reasons. This may be relevant e.g. in cases when the tanker uses LNG as fuel. We also would like to point out that, according to the data given in the IHS database, e.g. the average engine power (MCR) of tankers without an ice class built since 2000 is already quite close to the proposed new minimum engine power requirement (see section 6 of Annex to MEPC 71/INF.28) for this ship type, see the figure below. This indicates that there is very little margin to install less engine power in tankers.

9 MCR {kw] MEPC 72/INF.12 Annex 1, Page 7 Submitter Comments (with rationale) IMO minimum engine power (see MEPC 71/INF.28) and the average engine power of tankers without ice class built after the year P(min OW) MCR (OW) 0 Deadweight [t] This observation is not relevant for the ToR of this CG, but we would like to point out that the proposed minimum engine power for tankers with the range of deadweight from to about DWT seems to be higher than the current average installed MCR of this ship type. Another issue, which we think is important, is the design margin used by ship designers and shipyards when contracts are made for ships. Ship design is a complicated issue and these margins may be selected with regard to regulations (e.g. stability and damage stability regulations) or performance requirements of the ship (like the required deadweight (with regard to the lightweight

10 Annex 1, Page 8 Submitter Comments (with rationale) of the ships) as well as the required speed of the ship) given in the contract. For this reason, the attained EEDI values for phases 0 and 1 are often clearly below the required EEDI value. Therefore, careful analysis should be done in order not to impose too stringent requirements for the attained EEDI. 2. Fast ships The information given in the IMO data base indicates that gas carriers and containerships have already been built to meet phase 3 requirements for the required EEDI. Especially with regard to containerships, this trend is clearly evident. Quite probably the observed improvement of these ship types has been achieved by reducing the service speed of these ship types. For refrigerated cargo ships, no conclusions can be made due to very limited data in the IMO data base. More detailed analysis of the potential of the possibilities of this ship type to meet phase 3 requirements of EEDI taking into account application of the energy saving applicable to this ship type should be done. Germany Japan Netherlands With regard to ro-ro cargo ships (vehicle carrier) and ro-ro cargo ships the new reference lines for these ship types should be taken into account in the analysis, see Annex 15 to MEPC 71/17/Add.1. While analysis of database statistics allows for indication of the currently achieved EEDI level for the different ship types, it does not allow for prediction of achievable levels for the future. The degree of optimization of the ships is not known, therefore the potential for improvements is difficult to predict. Therefore, for each ship type, best performers should be selected in representative numbers for further analyses. In general, analyses should identify the reasons for their good performance, consider their safety and predict potential for further improvements. Japan is of the opinion that we need to collect ships attained EEDI data and the information of energy saving applied to the existing ships based on the EEDI database and other data, and to analyse further possible reduction of attained EEDI by applying additional/newly developed energy saving and other measures based on the latest development of energy saving. General remark: conclusions and forthcoming policy should be based on sound data. This means, among other things, there should be enough data (statistically wise). Due to the nature of calculation of the EEDI value and the physical design of a vessel, large vessels typically have a better correlation allowance to the EEDI reference line. With larger Froude numbers, it could be more convenient to meet the EEDI requirements than for smaller, shorter ships with a higher Froude number. As the scatter in the database on EEDI values is

11 Submitter Republic of Korea Russian Federation United Kingdom MEPC 72/INF.12 Annex 1, Page 9 Comments (with rationale) relatively high for smaller ships, care should be taken to assess the EEDI for the ships with a lower DWT. This is valid for all ship types. It is considered that appropriate evaluation for the EEDI data should be needed when reviewing beyond Phase 2 based on the vessels delivered to date, because the calculation of the EEDI has been revised very frequently. Therefore, the differences due to these revisions should be properly considered for EEDI review beyond phase 3, especially for B/C and Tanker, which EEDI and Minimum Propulsion Power are applied as mandatory. No comment It appears technology that can be qualified under the EEDI data base is only those affect calculation formula (electric and mechanical factors). It may worth exploring other contribution such as hydrodynamic, adding fins and changing ship s stern profile. Suggest to have another look at definition of innovation technology to expand the scope of coverage. United States Please see the attached slides with a suggested way of looking at the EEDI database information. Because the database is voluntary, it would be helpful to have more information about which ships are included in the EEDI database. Specifically, is data included for every ship built in each year, or only for those ships classed by specific classification societies. Also, if a significant number of ships elect not to participate, does this result in biased data (for example, ships that only just meet the requirements may elect to not provide the data). EC Further consideration should be given to how to ensure an unbiased and more complete data set to inform a discussion about any appropriate future changes to the reduction rates and effective dates. The EEDI requirements need to trigger innovation in terms of energy efficiency of ships. Therefore, it is crucial that existing ships and ship designs are required to be further improved. Phase 3 requirements need to be sufficiently ambitious to require such improvements. To allow for evidence-based decision making on the time period and reduction rates for phase 3 requirements as well as for possible phase 4 requirements, it is important to understand the performance of phase 0 and 1 ships, in particular the best performers (those with the lowest EEDI) and to understand the use of innovative to further improve energy efficiency of ships. For this purpose, statistical analyses on a per ship type basis should include but not be limited to: Average performance of 10% best EEDI required ships, as relative distance of attained EEDI to their respective reference line (in %) Percentage of ships already meeting phase 3 requirements

12 Annex 1, Page 10 Submitter CESA CSC Comments (with rationale) Use of innovative (4 th and 5 th term of the EEDI formula) as indicator for untapped improvement potentials of existing designs, expressed as percentages of ships with mandatory EEDI and as percentage of ships within the 10% bestperforming group These analyses could be carried out for all ships with mandatory EEDI and for phase 0 and 1 ships separately (if sufficient data available). Furthermore, for some ship types (e.g. bulk carriers and oil tankers) it might be necessary to differentiate into size categories. Simple example (for general cargo ships, EEDI database as of 1 August 2017): 32 ships with mandatory EEDI, 27 with EEDI requirement Average performance of 10% best ships: 57% better than reference line values 22 out of 27 ships meet phase 3 requirements (81%) Use of innovative : 0 out of 27 ships (0%) An in depth analysis of the EEDI database is difficult due to lack of parameters, e.g. speed or information about the innovative that have been utilized by some of ships included MEPC 71/INF.14, however, reveals even without additional parameters that many ship types already deliver significant reductions, sometimes even exceeding phase 3 requirements, although the improvement potential of alternative fuels and innovative has not or only marginally been utilized. Similar to design standards in other modes of transport, the logic behind EEDI is force industry to build more efficient ships. As it stands, EEDI is failing this task as ships with non-mandatory EEDI perform similarly to those that have mandatory EEDI according to IMO EEDI database. That means that, it is the prevailing market conditions rather than the regulation that drive design efficiency of newbuild ships. In general, IMO EEDI data base provides sufficient data and the following metrics/methods can be used to analyse the EEDI performance of new ships: 1. Average distance to reference line of 10% best performing ships in each ship class category. 2. Median distance to reference line of each ship class category. 3. Share of ships in each class category already meeting Phase III requirements (i.e. -30% compared to reference line). 4. Share of ships in each class category using innovative expressed in 4 th and 5 th terms of the EEDI formula. 5. Difference in energy efficiency performance of ships with and without 4 th /5 th term in each ship class category. An example for containerships with mandatory EEDI using the data base ending 1 August 2017: (a) A total of 258 ships in phase 0 and 1 of the EEDI; (b) Median performance (distance below reference line) of new ships: - 43% (c) Share of total ships meeting Phase III (-30%) requirement: 70% (d) Only 9% of containerships have reported the use of innovative under 4 th /5 th terms of the formula. Specifically, this equals to 24 out of 258 ships.

13 Annex 1, Page 11 Submitter IACS ICS Comments (with rationale) (e) Ships that have reported innovative under 4 th /5 th terms of the EEDI formula have on average 30% better EEDI than those not reporting. Specifically, the former have on average 8.9 g/tonne-mile EEDI vs g/tonne-mile EEDI of the latter. In the biggest ship size category (>200,000 dwt) this difference go up to 32%. 1) From the IMO EEDI database, it can be seen that innovative are not used 2) However, the technology that can be qualified with the EEDI database is limited to items that affect the calculation formula (electrical and mechanical factors) 3) Further optimisation of the existing technology (hydrodynamics, adding fins and change shapes in way of ship s stern) may have more contribution. Hull optimization is very important, but not listed as a technology; main dimensions, bow shape, aft ship, propeller & rudder characteristics can have a significant influence on overall fuel efficiency. 4) In this regard, not all the introduced in MEPC 68/INF.38 are qualified as innovative technology in the table 5) We may need to revisit the definition of innovative technology The principal concerns of ICS at this time are that: Maintaining the safety of ships must be the overriding priority; The actual performance of ships designed to achieve the necessary or improved EEDI phase targets should be considered, rather than only focusing on the attained EEDI values in the EEDI database; The analysis of further EEDI phase reductions considers what is practically achievable; To ensure that statistical outliers will be properly considered and addressed. When considering EEDI reductions, it is essential that ships still retain sufficient power to maneuver safely in sea conditions which they can expect to encounter in service. This is expanded upon in our response to question 1-3 but in short we believe that outstanding concerns over minimum power must be resolved before agreeing any further EEDI reductions or early implementation of EEDI phase 3. This necessitates agreement on the degree of ship control/maneuverability which a ship is to be able to maintain in adverse weather and the sea state which should be taken to represent these adverse weather conditions. If this is not done then ICS considers the risks of ships and lives being lost at sea and of major pollution incidents as a direct consequence of ships lacking sufficient power to maneuver safely and maintain control in adverse weather will be completely unacceptable. ICS would remind the group that the Maritime Safety Committee (MSC) is the IMO body responsible for decision making on matters of safety therefore there will need to be coordination between the Marine Environment Protection Committee (MEPC) and MSC, this is not simply an environment/marpol matter to be decided by MEPC.

14 Annex 1, Page 12 Submitter Comments (with rationale) The EEDI database in itself is not sufficient to undertake the required analysis for the EEDI review since it gives no indication on whether the ships concerned are able to maneuver and maintain appropriate levels of ship control in a range of weather conditions and sea states. Consideration of ship performance should be achieved by engaging with operators and reviewing in-service performance and capabilities of ships designed to achieve the necessary or improved EEDI phase targets. This is particularly pertinent to considering what weather conditions within which ships should be expected to be capable of maneuvering safely. The group should also recognize that industry is still gaining experience of the operational characteristics of ships designed to achieve the necessary or improved EEDI targets and that as such it is still an emerging area of knowledge. The analysis should only consider attained EEDI values along with in-service performance and ship safety and should not consider EIVs except for possible cases where there are no attained EEDI values to consider. This is to avoid erroneous conclusions derived from an analysis of estimated values which cannot be relied upon. There needs to be careful consideration of statistical outliers, both in terms of avoiding the correspondence group recommending requirements that will not be achievable for certain ships, and also to avoid statistical outliers distorting analysis for other ships. ICS considers that consideration of EEDI phase 3 needs to be evidence based and that it is essential for those actively involved in ship design and construction to provide information on what level of energy efficiency improvement can be achieved within the specified time frame. This is to mitigate the risk of agreeing to reduction targets based on academic studies which are not practically implementable. As ships increasingly adopt novel and increasingly complex systems to improve energy efficiency it is important to ensure that the claimed efficiency gains of such methods are representative of what the ship can be reasonably expected to achieve in service (see also comment to 2 3). Since a switch to LNG (or other alternative fuels such as methanol) may be necessary for some ship categories to achieve EEDI phase 3 then any deliberations by the group on early implementation of phase 3 must consider the availability of LNG (and other fuels which may be used to improve EEDI values) and whether the supply chain and industry is ready for such a change. In particular, this needs to consider that the ship types particularly affected, such as bulk carriers, often trade to ports in remote areas which do not have advanced supporting infrastructure and require high endurance. INTERFERRY The EEDI formulation is a statistical approach to establish mandatory requirements. In order to ensure meeting basic principles for statistical analysis, some acceptance criteria should be developed for significance, deviation and size distribution. If such criteria

15 Annex 1, Page 13 Submitter RINA Comments (with rationale) cannot be met, the further use of the statistical approach basing future requirements on historic correlation must be done very sensibly. Some technical improvement measures are well known, but their efficiency improvement may be very dependent on the individual ship design and great care must be taken not to presume that results from retro-fits by necessity correspond to efficiency improvement for new designs. RINA considers that the EEDI database may only be used to obtain an overview of historical to present achievement of the world fleet and does not contain sufficient information to make predictions on the future level of EEDI reduction phases. The most obvious approach likely to be taken will involve establishing the existing level of performance from the database and then to add some percentage allowance based on the latest information of energy saving to derive values for future phases. In RINA s opinion, this approach has several serious problems: 1. Such a technology list has a self-selection bias only owners or equipment manufacturers who have derived savings will provide information, however it is well known in design and operational circles that there are also many instances where the have failed to provide improvement, or that the improvement is extremely ship and operating route dependent. Such situations will not be reported to the list, and inherently skews the potential savings. Making a generalisation that all such technological applications will provide x% savings that can be summed up will very likely overstate the savings. 2. We do not have robust and standardised procedures in place for measuring and verifying savings that have been submitted, and in many cases shipowners also do not have this ability. 3. Some show large improvements in retrofit situations because the ship is no longer operating at its design condition; e.g. speed due to slow steaming, and the non-optimal hydrodynamics are being corrected by the technology. When applied to a newbuilding that is optimised to an operating profile, these improvements can rapidly diminish. 4. The ad-hoc nature of the EEDI database itself is of some concern as there is clearly some discrepancy between the numbers of ship in the database vs the numbers of ships delivered each year The EEDI framework encourages shipyards and designers to optimise performance to the EEDI point, sometimes to the detriment of real operational profiles particularly when some have a benefit over a very small range and may add resistance when operated outside of design point. RINA urges caution in attempting to set a higher level of reduction which may prove counterproductive in operational situations. RINA believes that one way to obtain a degree of certainty on possible future levels of reduction is to commission or request ship designs from shipyards, ship designers, towing tanks, etc. covering a range of ships and sizes that should be evaluated for efficiency, safety and operability.

16 Annex 1, Page 14 Submitter Comments (with rationale) RINA also notes that it is important that the correspondence group be made aware of 2 pieces of relevant regulation: 1. the unified interpretations contained within MEPC.1/Circ.795/Rev.2 which indicates that we will continue to take delivery of Phase 0 ships until 1 January 2019, and Phase 1 ships until 1 January 2024 depending on contract and keel laying dates 2. The minimum power lines for tankers and bulk carriers as contained in MEPC.1/Circ.850/Rev INTERIM GUIDELINES FOR DETERMINING MINIMUM PROPULSION POWER TO MAINTAIN THE MANOEUVRABILITY OF SHIPS IN ADVERSE CONDITIONS, AS AMENDED were increased in 2015 as compared to the previous revision agreed in RINA notes that based on a quick investigation using the IHS database, a large proportion of tankers delivered between 2013 and 2016 did not meet the revised minimum power requirements, while almost all met the earlier minimum power requirement. RINA also notes that the current guidelines only apply to Phase 0 and Phase 1 ships, and it is not known what level of minimum power will apply to future phases. Application of the revised guidelines will most likely result in increased EEDI scores across the board for tanker and bulk carriers unless some means of decoupling minimum power from the EEDI calculation may be found. WSC Finally, RINA would like to point out that future phases of EEDI need to be aligned with the general IMO GHG reduction strategy in order to avoid too many constraints on possible design and technological solutions. One suggestion is to look at number of ships that have utilized the 4 th and 5 th terms of the EEDI equation. This will give an indication to what extent new technology is implemented. A quick review according to the above shows that out of 2091 ships, only 1.3% have utilized the 4th term and 0-1% the 5th term. Utilization of these terms of the equation has only happened in the Vehicle Carrier and Container groups.

17 Annex 1, Page : It may be difficult for some ship type to make statistical analysis because of the lack of enough data in the EEDI database. How do you think which ship type has enough data and which ship type doesn t have? And if specific ship type does not have enough data, how the ship type should be evaluated/treated in this review? Submitter Australia Brazil China Denmark Finland Germany Japan Netherlands Comments (with rationale) No comment Brazil would like to point out the absence of data for very large bulk carriers bigger than dwt. It seems insufficient data is available for refrigerated cargo carriers, ro-ro cargo ships (vehicle carrier), ro-ro cargo ships, ro-ro passenger ships and LNG carriers. Referring to the comment below to 1-3 regarding bulk carriers there is only one observation on VLBC (very large bulk carrier) above DWT. It will be difficult to give a fair judgement on the problems to meet the tier 3 level on these big ships. However, as large bulk carriers have a great deal of similarities with tankers, the problems on large tankers, which is explained in the comment to 1-3, might also be present on large bulk carriers. It seems to be clear that, based on the current information in the IMO data base, insufficient data is available for refrigerated cargo carriers, ro-ro cargo ships (vehicle carrier) and ro-ro cargo ships. If the data base does not contain sufficient information, other means like case studies should be used for an estimation of future requirements. This might also encourage further entries to the database. Based on the number of data contained in the EEDI database, Japan is of the opinion that the analysis could be done for bulk carriers, tankers, container ships, gas carriers, general cargos and Ro-ro cargo ships (vehicle carrier). However, availability of the data depends on sizes of ships. For the other ship types, there is no/few data and it is difficult to consider the possible EEDI reduction rate. Thus, it is suitable to wait for a while until we have enough data for each of these ship types. When assessing the content of the EEDI database caution should not only be given to the amount of data per ship type. The distribution and scatter of data over the ship size (DWT) also has to be taken into account. For example, the general cargo ship database below 10,000 DWT is under-populated. Next to this it shows high scatter values. These ships do represent a significant part of the world fleet and especially the Netherlands fleet. As submitted in MEPC 64-4 (see below figure) there is a significant scatter in EEDI values for small ships.

18 Annex 1, Page 16 Submitter Comments (with rationale) Republic of Korea Attained EEDI (Phase 1) shows that the vessels in capacity range of 50,000 ~ 100,000 have more margin. While the vessels in capacity range of 50,000 ~ 100,000 meet phase 2 criteria already, the vessels in capacity over 300,000 do not meet phase 2 criteria. Please see below figure.

19 Annex 1, Page 17 Submitter Comments (with rationale) Figure : EEDI Database for tankers Suggestion: The reference line for high capacity range to be adjusted, so that the margin on each capacity is rational. Russian Federation Sweden Not enough data for the analysis of ships with high ice classes For Bulk carrier, Gas carrier, Tanker and Containership there seems to be enough data to make statistical analysis. The other ship type doesn t seem to have enough data. Ro-ro cargo and Ro-ro passenger ships should not be evaluated in this review, in part because there isn t enough data. But also because the new reference lines should be taken into account, see MEPC 71/17/Add.1 annex 15. Until we know the effect of the new reference line phase 3 should not be changed.

20 Annex 1, Page 18 Submitter Comments (with rationale) United Kingdom It is our view that file name Annex 1 EEDI data as of 1 st Aug 2017 provided by Co-Ordinator is good. But we welcome other contribution. United States The EEDI database contains a relatively large amount of EEDI data for only 4 ship types: bulk carriers (1,198 ships), tankers (692), containerships (360), gas carriers (160). There is a small amount of data for 4 ships types: general cargo (53 ships), refrigerated cargo (2 ships), RoRo vehicle (29 ships), and RoRo cargo (13 ships). There is no data for 4 ship types: combination carrier, LNG carrier, RoRo passenger, cruise ship w/nonconventional propulsion. For the 4 categories for which there is a relatively large amount of data, this data may be used as the basis of analysis to examine potential adjustments to some aspects of the program. For example, the data can be used to examine the Phase 3 effective dates, and specifically whether it is reasonable to pull ahead the Phase 3 requirements to an earlier effective date given so many ships currently achieve those reductions. However, there is not enough data available now to inform a discussion about revising the Phase 3 reduction rates or the potential for additional Phase 4 reductions. Specifically, there are a considerable number of vessel designs that do not meet the Phase 3 reductions, and we don t know why that is the case. EC CESA For those 8 ship categories for which there is no data or only very little data, it is not possible to develop any observations about experience of these ships with respect to achieving the EEDI. Without data, it is not possible to do statistical analysis about what they may be capable of achieving as a group. It may be reasonable to retain the Phase 3 reduction rates contained in Regulation 23 despite the lack of data in the EEDI database, since these reduction rates were agreed to at the time of the Annex VI amendments. However, without sufficient data it is not possible to know whether these ship types would be able to achieve earlier effective dates. In general, statistical analyses require minimum sample sizes of around 30 to 35 items. With the EEDI database as of 1 August 2017, this requirement is met for bulk carriers, gas carriers, tankers and container ships. For general cargo ships, the current sample can be considered as borderline case, but it can be expected that more data will become available before decisions on the phase 3 requirements will be taken. In the absence of sufficient EEDI data, all other ship types need to be addressed by using other information, including case studies of recent ship designs and energy efficiency. It can be noted that the ship types concerned are not the most relevant ones in terms of GHG emissions (in total responsible for less than 20% of emissions, according to 3 rd IMO GHG study). It is obvious that for some ship types no or little statistical data is available. The lack of data, however, is no proof that the above observation is not also valid for them. The principal reduction potential of the individual is available to all ship types and also the ship types with little or no data offer numerous opportunities to combine conventional and innovative in a complimentary and favourable manner.

21 Annex 1, Page 19 Submitter CSC IACS ICS Comments (with rationale) Up to August 1, 2017, EEDI data base contains enough information to analyse the performance of all major ship types. Notably, there is a good sample of ships in the category of bulkers, tankers, container ships and gas carriers. Results presented in section 1-3 below provide evidence to significant over-compliance with little innovative technological uptake. The sample size for general cargo ships is comparatively small, 32 ships under the Phase 0 and 1. However, the similar performance of the obtained EEDIs of non-mandatory ships (in total 11) provide additional information on attainable EEDI potential of the general cargo fleet. In general, according to the IMO database, the median performance of general cargo ships is above 50% below the respective reference line. In addition, according to a recent study of EEDI of more than 1500 general cargo ships built between based on EIV values (and taking into account the empirical relation that the EEDI is about 10% lower than the EIV), on average 45% of ships have EEDI at least 30% below the reference line. This analysis of the EEDI values encompassed around 324 ships, which represent a solid sample for statistical analysis. Additionally, even the study used EIV, a simplified formula for calculating design efficiency, the conclusions of the study in other ship class categories correlate well with data of official EEDI scores released by the IMO secretariat as part of this CG. The information provided by the Co-ordinator is a good reflection of the reality. (file name Annex 1 EEDI data as of 1st of Aug 2017) Currently ICS believes that the limited amount of data for attained EEDI of ships in the following categories may make it difficult to properly analyze statistics: General cargo ship Refrigerated cargo carrier Combination carrier LNG carrier Ro-ro cargo ship (vehicle carrier) Ro-ro cargo ship Ro-ro passenger ship Cruise passenger ship having non-conventional propulsion In the case of LNG carriers it is recognized that these ships are included within the gas carrier category for phase 0, however there is a shortage of data for large gas carriers above 60,000DWT. ICS is also concerned that different categories of gas carriers may have very different auxiliary loads depending on cargo refrigeration requirements and that this needs to be considered.

22 Annex 1, Page 20 Submitter Comments (with rationale) ICS would emphasize again that statistics and data are insufficient to analyze whether ship designs are capable of meeting the phase 3 criteria safely and reliably unless supported by an evaluation of in-service performance and maintaining sufficient minimum power to ensure the safety of the ship. INTERCARGO By looking at the total number of Bulk Carriers it would initially appear that 1198 total number of ships would provide enough data to perform a statistical analysis. However, when the number of ships is broken down into the vessel sizes this is not the case. For example if we look at the Very Large Ore Carriers data is only available for 15 vessels, thus for some vessel types/size more data is needed. INTERFERRY When establishing of the EEDI reference lines for ro-ro cargo and ro-ro passenger ships, for both sub-categories, there was only data available for around 100 ships each, out of a population of some 500 ships in each sub-category. During the 2016 EEDI review, it became apparent that the available data did not represent the characteristics of the fleets well enough, which prompted a significant correction to be adopted by MEPC71. The current EEDI Database contains only 13 ro-ro cargo ships and no ro-ro passenger ships at all. MEPC71 also agreed that an upper size threshold (17,500 DWT) had to be introduced for ro-ro cargo ships, since the reference line was populated by relatively small ro-ro cargo ships. When reviewing the 13 ro-ro cargo ships currently included, we note that only 3 are smaller than 20,000 DWT, which would indicate that most of the ships included are not representative for the typical ro-ro cargo ship design. As under request 1-1, we would argue that acceptance criteria for statistical relevance should be agreed before any recommendation can be made for a ship category. RINA RINA considers that analysis of the data is problematic not because of a lack of a statistically significant number of ships (though that may be the case for some classes of vessels as a whole, or within certain size ranges), but because there is insufficient information about each ship to draw concrete conclusions. EEDI scores of ships may be skewed by the application of correction factors for ice class, voluntary structural enhancement, cranes, etc, but this data is not provided in the database. Information relating to main dimensions, EEDI speed and Pme are also missing. Since the IMO numbers are not disclosed, RINA believes that anonymity will be maintained even if all these parameters are disclosed, and at the same time there could be much more valuable and accurate dissemination of the state of the art. Crucial information about application of or fuel is also not covered by the database. Using the categorization in MEPC.1/Circ.815, we arrive at the following table: Technology Hull Optimisation Cat A (Hydrodynamic) x Cat B (5th term) Cat C (4th term) Notes

23 Annex 1, Page 21 Submitter Superstructure Optimisation Propeller Optimisation Podded Drives Contra Rotating Propeller Vane Wheel Ducted Propeller Pre-Swirl Device Post Swirl Device Twin Skeg Waste Heat Recovery Hybrid Propulsion x x x x x x x x x Comments (with rationale) x Diesel electric ships would not be subject to EEDI, so would not be included in database Fuel Cell?? Could be Cat B or C Coatings x Air Lubrication x LNG Represented in the carbon conversion factor Cf Biofuels Represented in the carbon conversion factor Cf Wind power x Solar power x Wave power x Optimisation of dimensions x Lightweight reduction x Hull vane x

24 Annex 1, Page 22 Submitter Technologies not explicitly in Annex 3-2 Batteries LED Lighting Variable Speed Motors Comments (with rationale) x x No provision for treatment of batteries for peak shaving in the EEDI framework In theory this is category C, but although it is widely installed, it is probably not used in PAEeff In theory this is category C, but although it is widely installed, it is probably not used in PAEeff So out of the 23 provided for in Annex 3-2, 14 are hydrodynamic in nature and uptake of these innovative cannot be derived from the EEDI database. Fuels also cannot be deduced from the database because the Cf factor is not recorded, and finally hybrid propulsion that utilizes a diesel electric system would not appear in the databases at all, unless they are cruise ships. So in effect the database provides us with no information for about 74% of the innovative that could be applied. In terms of enabling uptake of, batteries, assuming they are used to provide electrical power to auxiliaries are not well catered for since there are no clear guidelines for their use. Uptake of LED lights and variable speed motors may also be significant, but although these are technically category C, designs at the moment do not need to rely on these to achieve compliance with EEDI and therefore their use may not be recorded. With regards to in general, RINA would suggest that we use technology readiness levels (TRLs) to categorise the availability of in order to introduce more objectivity into the analysis. WSC The EEDI data for Refrigerated Cargo Ships, Ro-Ro cargo Ships (Vehicle Carrier) and Ro-Ro Cargo Ships is insufficient for any sound statistical analysis. Further evaluation of these groups should be postponed until further data is available.

25 Annex 1, Page :Given the information contained in the EEDI database, what is your preliminary assessment on whether current ship designs are capable of meeting the phase 3 criteria or not? Are there any ship type or ship size which may not meet phase 3 criteria? The views could be given to each ship type or ship size. Submitter Australia Brazil China Comments (with rationale) Based on current data, it does appear that there are a number of vessels in the database provided which meet or exceed the Phase 3 reference line currently. On this basis, it is possible that with improvements in the ability to meet or exceed the reference lines will also improve. The absence of data doesn t allow a preliminary assessment on whether there is a problem of implementation of EEDI for large bulk carriers, a matter that has been brought up to MEPC a few times. However, we have received information from a Brazilian shipowner that these vessels may find problems to meet phase 3 requirements with the current available technology. According to the feedback of our industry, using current mature energy saving technology and considering minimum power requirements, it s almost impossible to meet Phase 3 requirements for majority of large bulkers and large tankers. Denmark A surprising phenomenon which could be seen on figure 2, 3 and 4 above is the relative unchanged level of EEDI from phase 0 and phase 1. An explanation, although based on very few phase 1 data, could be that the EEDI reduction has already been made on ships in phase 0 by reduction methods already available on the market. This includes: speed reduction to reduce propulsion power reduced SFC (Specific Fuel Consumption) obtained by engine methods as o derating (power and engine speed) o change of engine type parameters (bore/stroke ratio) o optimizing engine tuning profile. increased propeller efficiency obtained by reduced engine speed and increased propeller diameter increased hull efficiency obtained by change of dimensional parameters, modern hull painting etc. It is worth mentioning that the reduction on SFC from the reference value of 190 g/kwh to the commonly obtained value below 170 g/kwh on two stroke engines are obtained by several engine efficiency reduction methods in the phase 0 period. This increased efficiency contributes with an EEDI reduction of around 10%. Regarding Containerships, it is noted that the larger vessels have obtained a remarkable 40-50% EEDI reduction, while the smaller ships although still considerable has obtained a 15-40% reduction. This impressing reduction is probably based on the comprehensive speed reduction on these ships, which has been possible due to the traditional high speed on the large containerships (24-26 knots on post Panamax ships).

26 Annex 1, Page 24 Submitter Comments (with rationale) Based on the observations it seems possible in general to meet phase 3 for Containerships. However, on smaller container ships below DWT, which are normally feeder ships with a lower speed, it might be difficult to meet the requirements, although not impossible. The speed of container ships, built in the period , are shown in fig. 5 for reference. Regarding Tankers, it is easy to see that the EEDI obtained in phase 1 coincides with the values of phase 0. Furthermore, and this is an important point mentioned by several experts, it seems to be difficult to obtain a sufficient EEDI reduction on the large tankers (VLCC and ULCC). There are several reasons for this, but a simple explanation is given by the power and related speed on these ships, which is so low, that a further reduction does not seem to be fair and safe. The observed problems on achieving a satisfactory minimum power in adverse weather should be kept in mind. Even meeting the phase 2 requirement of 20% reduction seems to be a problem. An important relevant physical explanation on the difficulties to further reduce the power is related to the low speed/length ratio on these large ships. In this case the ship resistance is dominated by the wet surface resistance (related to speed^2) and not by the wave-making resistance, which has a higher exponent than 2. A speed reduction on large ships will accordingly give a relative low impact compared to speed reduction on smaller ships operating at a higher S/L ratio. Based on these observations it seems possible that phase 3 could be met on Tankers below DWT. On tankers above DWT it seems difficult to meet phase 3 and a fair limitation on the reduction rate should be considered, not only to keep a sufficient speed, but also to keep a safe power in adverse weather. The speed on tankers in the period are shown fig 6. Regarding Bulk Carriers the same observations as seen on tankers could be found, although large bulk carriers above DWT are only represented by one ship. The same problems found on large tankers will apply for large bulk carriers as well. The ship speed on Bulk Carriers, shown in figure 7, are slightly lower than the speed on tankers, around 0.5 knot, and a speed reduction would be more severe for this sector. Based on the figures and the close relation to tankers, it seems that bulk carriers will have some difficulties to meet phase 3 requirements even at low DWT, and the large bulk carriers will certainly have difficulties to meet the requirements as observed on large tankers. A limitation on the phase 3 reduction rate should be considered, especially on large bulk carriers in line with the observations made on large tankers, but might also be considered on bulk carriers in general.

27 Annex 1, Page 25 Submitter Comments (with rationale) Fig 5 Ship speed on Container ships observed in IHS database

28 Annex 1, Page 26 Submitter Comments (with rationale) Fig 6 Ship speed on Tankers observed in IHS database

29 Annex 1, Page 27 Submitter Comments (with rationale) Fig 7 Ship speed on Bulk carriers observed in IHS database Finland Based on the information given in Annex 1, it seems that it is already possible to build gas carriers, containerships and general cargo ships to meet the phase 3 requirements. With regard to bulk carriers, tankers, refrigerated cargo carriers, ro-ro cargo ships (vehicle carrier) and ro-ro cargo ships more detailed analysis should be done before any conclusions can be made. Due to lack of data, more detailed analysis should be

30 Annex 1, Page 28 Submitter Germany Japan Comments (with rationale) done also concerning combination carriers, LNG carriers, ro-ro passenger ships and cruise passenger ships having nonconventional propulsion. See also our reply to question 1-1. Most important question, to be answered by CG. Obviously, based on current ship design there are certain ship types that will be able to meet phase 3 criteria (container ships, general cargo ships) while others may not (bulkers, gas carriers). At the same time it is understood that the uptake of innovative technology for improvements of energy efficiency is currently quite low. Use of such will enable improvements for currently bad performing ship types. Before getting into the possibility of meeting Phase 3 criteria, Japan would like to explain our view on the progress of improving energy efficiency of ships. Recent newbuilding ships have improved their energy efficiency by utilizing various possible energy saving. Thus, energy efficiency of typical cargo ships such as bulk carriers and tankers has been considerably improved by applying almost all the available energy saving such as optimization of hull form and applying additional energy saving devices. On the other hand, there is no clear perspective or evidence that new energy saving /devices which could substantially further improve energy efficiency of ships would be applicable in the coming several years. Besides, although ships using LNG fuel may drastically improve attained EEDI, introduction of LNG fuels requires millions to tens of millions of dollars of additional cost for each ships. Furthermore, shore-based infrastructures for providing LNG fuel for ships are also necessary, and they will not be provided enough in near future like the implementation date of Phase 3. Thus, it is impossible to improve energy efficiency of all the ships by LNG fuels. The next possible measure is reduction of ships design speed for improving the attained EEDI. To improve the attained EEDI through reduction of the design speed, it is necessary to reduce the power of the installed engine. This may result in failing to meet the minimum power requirement especially for relatively slower ships such as bulkers and tankers. Besides, it is also necessary to consider the impact of lower design speed on shipping market and the capacity of global seaborne trade. Based on this situation of recent energy efficient ships mentioned above, Japan would like to explain its view on the possibility of meeting Phase 3 criteria. To improve the attained EEDI for ships, it is necessary for shipbuilders to have enough time for improving ships design and/or applying new energy saving. Early implementation of further reduction of EEDI value would cause serious burden or confusion to the shipbuilders, relevant manufacturers and whole the newbuilding market of ships because they need enough lead time for changing their design to meet new EEDI requirement and they would have difficulty in constructing series of ships

31 Annex 1, Page 29 Submitter Comments (with rationale) due to the change of the requirement. Therefore, the application date of EEDI Phase 3 requirement should be maintained as It is necessary to make detailed analysis based on the recent trend of developing new energy saving and the other available information for considering the applicability of Phase 3 criteria. As an initial comment on this request, Japan has the following views for each ship type and each ship size. All the data contained in the following comments are from the EEDI database. Regarding bulk carriers, there are few ships with which EEDI reduction rates are more than 30 percent. The EEDI reduction rates of many of handy bulkers, handy-max bulkers and Panamax bulkers are about 15 to 20 percent. Even if these ships could introduce and apply additional energy saving, in order to meet 30 percent reduction of attained EEDI, it might be necessary to reduce their design speeds. They might also not meet 30 percent reduction of attained EEDI by reducing design speed because of the minimum power requirement. The EEDI reduction rates of Cape size bulkers and larger bulkers are currently about 15 percent, and they might not meet 30 percent reduction of attained EEDI even if both applying additional energy saving /devices and reducing design speed of ships were possible. Regarding tankers, there are part of ships with which EEDI reduction rates are more than 30 percent. The EEDI reduction rates of many of tankers less than 50,000DWT are about 20 percent. Even if these ships could introduce and apply additional energy saving, in order to meet 30 percent reduction of attained EEDI, it might be necessary to reduce their design speeds. They might also not meet 30 percent reduction of EEDI by reducing design speed because of the minimum power requirement. The EEDI reduction rates of VLCCs are currently about 15 percent, and they might not meet 30 percent reduction of attained EEDI even if both applying energy saving /devices and reducing design speed of ships were possible. Regarding container ships, ships less than or around 20,000 DWT are still about 10 to 20 percent reduction of attained EEDI. Thus, even if these ships could fit additional energy saving, it is still necessary to consider carefully about the possibility of meeting 30 percent reduction of attained EEDI by applying additional energy saving. On the other hand, many of larger ships more than 40,000 DWT have achieved 30 percent reduction of attained EEDI. However, the major reason of this could be the considerable level of reduction of ships design speed of recent ships compared with the ships used for developing reference line, and thus careful consideration is necessary on this point. Regarding gas carriers and Ro-ro cargo ships (vehicle carrier), a part of ships are above 30 percent reduction of attained EEDI, but there are many ships which are about 10 to 20 percent reduction of attained EEDI. Even if these ships could introduce and apply additional energy saving, in order to meet 30 percent reduction of attained EEDI, it might be necessary to reduce their design speeds.

32 Annex 1, Page 30 Submitter Netherlands Norway Republic of Korea Russian Federation Spain Sweden See 1-2. United Kingdom Comments (with rationale) When installing less power it will be possible to comply for all ship types. This is why the minimum power issue is of utmost importance. Furthermore it seems generally recognized that some ship types (e.g. containerships, general cargo ships) are complying phase 3 without any innovative energy saving or design improvement. Purpose of the EEDI is to improve the energy efficiency of ships and to push the industry towards more energy efficiency. At this moment it seems that, at least for some ship types, ships can comply with phase 3 easily. As stressed already it should be kept in mind that there are differences (in distribution and scatter) within a certain ship type when the ship size is taken into account. In general, it seems like the phase 3 requirements can be met for container ships, and possibly gas carriers and general cargo ships. If meeting future requirements for a ship type will depend on size, we should consider separate requirements for various size segments. Also, when discussing a possible early implementation of a phase 3, reduction rates could be reduced to e.g. 25% for those ships that cannot meet the current reduction rate for phase 3 of 30%. If reducing the phase 3 reduction rate, a phase 4 needs to be established in order not to reduce the overall ambitions of regulation 21. It will be hard to meet the phase 3 required EEDI for slower full block ships such as Tankers and bulk carrier. This is not only because of the design aspect but also the requirement of minimum propulsion power. Additionally to meet the phase 3 criteria may also be difficult to the LPG and PCTC (Ro-ro cargo ship (vehicle carrier)). No comment Spain is of the view that all ships can meet the phase 3 if using innovative, with special treatment for bulk-carriers and possible very small container ships. We also consider that in case of early implementation the coefficients can be kept as such and phase 4 should increase the coefficients. We don t find substantial problems that could not be overcome if Phase 3 was advanced to Tankers and bulk carriers might have to apply innovation for meeting phase 3. Other ships type tends to be able to early compliance. It should bear in mind we might also take minimum power into consideration once the IMO guidelines is finalized (At the moment, only IMO interim guidelines is available). United States The Phase 2 reduction is 20%, effective ; the Phase 3 reduction is 30%, effective Only for the containership category are there a substantial number of ships achieving the 30% reduction ahead of time, and even for that category there are still many ships that do not currently achieve a 30% reduction. The clear majority of the ships included in the EEDI database currently do not achieve the Phase 3 reduction rates. Even though some of the data in the EEDI dataset suggest that at least some ships can achieve the Phase 3 reduction rates, the majority clearly do not. Further, the data clearly show that, even for those ships where there is a good amount of data, there

33 Annex 1, Page 31 Submitter EC Comments (with rationale) is much variation on the ability to do so both within and among ship categories. This is true even for containerships. The average reduction rate for container ships is about 37%. However, only the largest containerships (>100,000 DTW) have the most efficient designs (40% reduction on average). While containerships <100,000 DWT on average achieve the 30% Phase 3 reduction requirement, not all of them do and there is a significant degree of variation (+/- 12%) for these ships. Therefore, careful consideration should be given to the constraints experienced by containerships <100,000 DWT before any changes to the reduction rates are considered. For all ship types, advancing the Phase 3 effective dates could be considered as an alternative to more stringent reduction rates. This is because the data support the notion that ship designs can be made to achieve the Phase 3 reductions even if most designs are not doing so at this time. Preliminary analyses show the following results: Ship type Number phase 0 & 1 ships (mandatory EEDI) Average performance of 10% best (in % below ref line) 'Early compliance' phase 3 Use of innovative Bulk carrier % <1% 0% Gas carrier % 18% 0% Tanker % 33% 0% Container ship % 80% 10% General cargo ship 32 57% 81% 0% From this preliminary analysis it can be concluded that best-performers among phase 0 and 1 gas carriers, container ships and general cargo ship easily fulfil phase 3 requirements, with a very limited (only some container ships) use of innovative. Furthermore, best-performing phase 0 and 1 tankers also fulfil phase 3 requirements, without any reported use of innovative. For bulk carriers, it is noted that today's best performers (expressed as average of the 10% best to level out the impacts of possible outliers) are close to phase 3 requirements, without any reported use of innovative. By using innovative and/ or improved ship design, it can be expected that bulk carriers will be able to fulfil phase 3 requirements. As indicated in our reply to request 1-1, further, more detailed analyses might be necessary to differentiate into size categories. As intended, the EEDI should trigger improvements of ship designs and the use of innovative to increase energy efficiency. Having this in mind, phase 3 requirements for the above mentioned ship types should be advanced to For some

34 Annex 1, Page 32 Submitter CESA CSC Comments (with rationale) ship types strengthening of the phase 3 requirements would be needed to maintain an incentive for further innovation: gas carriers, tanker, container ship and general cargo ship. Based on review of available, their individual reduction potential and the possibility to combine conventional with alternative fuels and innovative phase 3 requirements can in principle be met by all ship types in In general, some efficiency targets, like the Japanese Top Runner Program, analyse the existing achievements on the basis and future targets on the basis of both the efficiency of the best-in-class in the current market and an analysis of the impact of innovations (METI, 2015). Taking inspiration of this highly efficient method, the EEDI achievements of the best of the class ships can be analysed. In addition, it is also pertinent to analyse the share of ships that meet and exceed phase III (-30%) target, as it provides a good indication of the efficiency penetrated into the fleet. The results indicate that tankers, gas carriers, containerships, and general cargo ships already comply with phase III requirements (10 years ahead of the required date); despite this over-compliance, only small number of containerships 9% - have reported using mechanical or electrical innovative energy saving covered under the 4 th /5 th terms of the EEDI formula. Among bulk carriers, on the other hand, less than 1% of the fleet has reached the Phase III requirement. However, the performance of the 10% best ships is not far away from achieving the phase III target (27%). Given that bulk carriers have not reported any energy saving under the 4 th /5 th terms of the EEDI formula, by resorting to those it can be assumed that ships in this class too can meet and exceed the Phase III target in the coming years. Given that these 5 ship types are responsible for the majority of global emissions, it would be worth to consider prioritising and expediting the revision of EEDI for these ship types before discussions on other ships (i.e. RoRo, ice-class, etc.). Ship type # of Ships Years (all ships in mandatory phases) Average distance to reference line of attained EEDIs of 10% best performing ships Share of ships meeting and exceeding Phase III (-30%) target Use of innovative (4 th /5 th terms) Difference in efficiency between ships with and without 4 th /5 th terms Share of global emissions (IMO 3 rd GHG study) Bulk carrier % 0.4% 0% n/a 18% Tanker % 26% 0% n/a 14% Gas carrier % 13% 0% n/a 5% Container ship % 71% 9% 30% 22% General cargo ship 32 57% 69% 0% n/a 7%

35 Annex 1, Page 33 Submitter Comments (with rationale) It is important to note that, these low EEDIs have been achieved with minimum (less than 10%) or no reduction in the installed main engine power, while design speeds and DWT capacities have remained largely stable or increased in some cases. This suggests better hull and propeller efficiency requiring less power to maintain the same speed. For example, the figure 1 below indicates that in bulk carriers segment main engine power was reduced within less 10% range since 2009 a hypothetical baseline year before the EEDI requirement, while design speed remained more or less the same or even increased in some cases. Bulkers in the biggest size category of 250, ,000 DWT have even increased their main engine power. In general, similar trend is also observed in tanker segment, too (Fig 2). This suggests that reduction in the main engine power is not the only way to achieve better EEDI and that observed limited reductions in the main engine power could rather be justified with better hull and propeller efficiencies rather than required EEDI per se. Given that the analysis underpinning Fig 1 and 2 also include non-eedi ships, this would additionally corroborate the assumptions that changes in the main engine power were rather driven by the prevailing market conditions as opposed to EEDI regulation. Figure 1: Development of engine power, size and speed of Bulk carriers (2009=100% engine power, i.e. baseline)

36 Annex 1, Page 34 Submitter Comments (with rationale) Source: CE Delft Figure 2: Development in engine power, size and speed of Tankers (2009=100% engine power, i.e. baseline)

37 Annex 1, Page 35 Submitter Comments (with rationale) Source: CE Delft IACS 1) Tanker and bulk carriers are facing difficulties. Whilst some tankers comply with Phase 3 this is not the case for bulk carriers 2) This may be, because these ship types are subject to the minimum power guidelines whilst others are not.

38 Annex 1, Page 36 Submitter ICS Comments (with rationale) 3) EEDI requirements (if further strengthened) cannot be discussed without considering minimum power requirements. IACS considers that this is quite critical, especially for the slower full form ships. 4) The attained EEDI of container ships lies by and large below Phase 3, the exception being feeder vessels (about 45% were below Phase 3 in ). The average value of attained EEDI for vessels > 4000 TEU delivered in 2016 is about 20% below Phase 3. ICS considers that this is not the most appropriate question and that we should actually be asking whether current ship designs are capable of meeting the EEDI phase 3 criteria safely and reliably. ICS does not believe that at this point there is sufficient data and operational experience to state that current ship designs in all categories can achieve EEDI phase 3 safely and reliably. Efficiency can be improved with relative ease simply by reducing the speed of ships and if the question of minimum safe power is ignored then a ship can be made to meet EEDI phase 3. However, as ships are provided with smaller engines because their service speeds are reduced then this introduces a conflict between efficiency and safety as the capability to maneuver safely in adverse weather or areas of strong current is reduced. This loss of maneuvering capability introduces a risk of ship collision and allision and of ships running aground and causing pollution. Therefore, ICS believes that this work must recognize this conflict and agree that maintaining safety of ships is paramount. Using the current guidelines for minimum power some ship yards have advised that for some categories of ship, such as bulk carriers, it may be necessary to switch to LNG fuel in order to achieve EEDI phase 3 to try and meet the requirements of both the EEDI and maintain sufficient power. ICS would draw attention to the fact that work to develop new minimum guidelines for minimum power is incomplete and that these new guidelines (MEPC.71/Inf.28) were not considered to be mature enough for completion at MEPC.71. The fact that certain categories of ship, such as container ships, are already achieving reductions exceeding the phase 3 target should not be interpreted as evidence that phase 3 is too conservative. For some categories, particularly bulk carriers, achieving phase 3 without compromising ship safety will be very challenging. Whilst switching to LNG fuel will assist these ships to achieve phase 3 it will not alleviate concerns about safety and minimum power. ICS is aware that some have argued that reducing engine power and service speed is not a requirement of the EEDI and that other efficiency enhancing measures are available. Since the biggest gains and lowest cost means to enhance EEDI values is to reduce speed and power it should be expected that this will remain the first option for ship designers therefore the issue of maintaining sufficient power to maintain safe maneuverability remains pertinent.

39 Annex 1, Page 37 Submitter Comments (with rationale) For all of the above reasons, ICS believes that whilst consideration the capability of ships to meet the EEDI phase 3 criteria may proceed in parallel with a review of minimum power requirements, no decision on early implementation of EEDI phase 3 or EEDI phase 3 reduction rates should be taken until issues around minimum power have been resolved. Two questions are central to any consideration of minimum power and maintaining maneuverability in adverse weather conditions: 1. What sea state should be adopted in order to quantify the adverse weather conditions to be used when establishing minimum power requirements? 2. What level of manoeuvrability/ship control should a ship be able to maintain in this sea state? Unless answers to the above two questions can be agreed then we will not be able to establish minimum power guidelines for ships. The adverse weather conditions used by the proposed draft guidelines (MEPC.71/INF.28) are considered to be too conservative by ICS and not representative of conditions which ships face in normal service. Additionally, the degree of maneuverability proposed in these draft guidelines is considered by ICS to be unacceptably low, based on an advance speed of 2 knots and considering maneuvering in coastal waters. ICS is aware that the 2 knot advance speed may be increased to a 4 knot advance speed but would still consider this to be inadequate under the weather conditions defined in the draft guidelines at this time. ICS also considers that deep ocean conditions should be considered since a loss of power and the resultant risk to crews and risk of collision is as unacceptable in deep ocean waters as in restricted to coastal waters. ICS is aware of a number of significant operational issues and technical deficiencies which have been experienced by ships designed to achieve the necessary or improved EEDI phase targets, including: stern bush and shaft damage caused by high deflection and bending moments as a result of installing heavy, large diameter propellers in combination with long stroke engines running at low speeds; high shaft fatigue levels which will cause premature crankshaft failure as a result of the engine taking excessive times to accelerate through the barred speed range; poor manoeuvrability in adverse weather or in areas with strong currents; loss of power as a result of the engines having insufficient torque in areas with strong currents. The above problems could lead to a loss of propulsive power at a critical moment and as such have the potential to result in the loss of a ship along with the crew and/or a major pollution incident which could cause major environmental damage. Although ships are required to satisfy the severe wind and rolling criteria of the intact stability code in a dead ship condition, a loss of power

40 Annex 1, Page 38 Submitter Comments (with rationale) clearly increases the risks of collision and allusion or of going aground and the risks to crews from excessive rolling in the event of a loss of power may be severe. INTERCARGO Looking at the provided data, it is clear that Bulk Carriers will face difficulties in reaching Phase 3 as only approximately 0.5% of bulk carriers are currently meeting Phase 3 requirements. Of the 1193 Bulk Carriers less than 50% meet the phase 2 requirement, however if this is broken down into vessel size i.e. Handysize, Handymax, Supramax, Panamax, Capesize, VLOC) then for certain vessel sizes the percentage meeting Phase 2 is significantly less than 50%, which further indicates that bulk carriers, or at least certain size of bulk carriers, will find it extremely challenging to meet phase 3 criteria. In addition to the data providing clear evidence that bulk carriers will find it difficult to achieve phase 3, INTERCARGO is of the strong opinion that the safety aspects cannot be ignored. Before any decision is made regarding Phase 3, the minimum powering requirements for bulk carriers need to be very carefully considered. INTERFERRY RINA For the ro-ro cargo and the ro-ro passenger ships, INTERFERRY believes that these ship types should be able to meet the Phase 2 criteria after the recent correction of the reference lines. Having said that, given the bespoke nature of the ro-ro ship types, until a reasonable number (to be defined) of real ship designs have been developed and for which the attained EEDIvalues are known, it would only be guesswork whether Phase 3 can be met. RINA s opinion is that there is no statistical basis for EEDI review for the following ship types due to lack of ships compared with the corresponding size of the world fleet the presented data cannot be considered to be representative: Refrigerated cargo carrier (it should be noted that this type of vessel is undergoing major changes from palletized to containerized) Combination carrier LNG Carrier Ro-ro cargo ship (vehicle carrier) Ro-ro cargo ship Ro-ro passenger ship Cruise passenger ship having non-conventional propulsion RINA believes that for the following types of ships the statistical data seems sufficient, but the lack of information contained within the EEDI database means we cannot make predictions about whether Phase 3 and beyond is achievable, also noting our earlier

41 Annex 1, Page 39 Submitter Comments (with rationale) comment regarding the effect of the increase in minimum power requirements which has not manifested itself in the database yet: Bulk carriers Gas carriers Tankers Where there are statistical outliers such as for tankers, gas carriers, general cargo ships, we would suggest that the responsible Class Society carry out detailed analysis of why such scores were achieved and disseminate that information. It is important for the industry to understand as a whole whether these values are sustainable and accurate, or the result of correction factors. WSC For containerships, it is clear that Phase 3 is already being achieved in the majority of cases, however before we set a level for future phases, RINA would like to see some detailed analysis about how these values have been achieved, perhaps in relation to the described in Annex 3-2, and why some others have not been able to achieve Phase 3 levels of reduction. All ships, except for Ro-Ro Cargo ships, can meet the Phase three requirements by reducing installed power and hence speed. As commented earlier though in the previous CG on EEDI, reduction of installed power and speed reduction should not be considered as technological developments for improved energy efficiency. Such speed reduction reduces transport work. To be considered energy efficient you should compare energy requirements for the same amount of transport work.

42 Annex 1, Page 40 Request 2-1 : Members are invited to provide or update the following information of each technology contained in MEPC 68/INF.38. Request 2-2 : Members are requested to provide the information of newly developed energy saving which are not contained in MEPC 68/INF.38. When providing the information about newly developed energy efficient devices, following information should be included, if available. Written comments which are not directly related to individual energy saving for EEDI Submitter Denmark Information Thermal Energy Efficiency Design Index (TEEDI) Currently the thermal efficiency of a ship is not taken into consideration. We believe it is possible to achieve substantial fuel savings if 1) the thermal energy system was optimised during the design process, but also 2) the thermal efficiency in relation to cargo handling. EEDI does not consider cargo for good reason. Nevertheless, we are of the view that a consideration of the CO2 emission per ton steam or thermal oil against an index (or simply just an efficiency demand) would be sufficient, no need to relate it to ship capacity or speed. This will force boiler manufacturers to focus to a higher extent on fuel efficiency. Increased utilisation of the thermal energy available from propulsion and auxiliary engine operation dependent of ship type, could reduce the need for electrical heaters (pools, saunas, hot water boosters etc.), electrical pumps, compressors and air handling units.consideration of the thermal energy system needs to be against a proper index as we know it from EEDI, because the engine efficiency matters (the better engine efficiency the less available thermal energy, and the less critical it is to utilize this). The index should be built upon waste heat emitted to atmosphere or dumped to sea per ship capacity and speed. EEDI penalizes combustion units connected to Exhaust Gas Cleaning Systems (EGCS) Today shipowners need to choose a sulphur compliance strategy, which amongst other is the choice between installing an EGCS or using compliant fuel (petroleum based). If a shipowner chooses an EGCS as his sulphur compliance strategy, he will be penalized on his EEDI calculation, due to the higher CO2 emission per heating value of non-compliant fuel vs. compliant fuel. The paradox in this is, in order to reduce the sulphur content during the refinery process, the refineries need to utilize energy consuming processes like hydrocracking, resulting in higher overall CO2 emissions. The estimated increase in CO2 emissions varies from author to author. The EnSys/Navigistics fuel availability study, expects it to be up to 4%*, but a paper from Transoleum, which speculates in another preferred refinery strategy (increase crude oil throughput, instead of increase in hydrocracking capacity), expects CO2 emissions to increase by 47%.** The truth is found somewhere in-between; we often hear specialists estimating it to be between 10-15%. Exhaust gas cleaning is not less valuable than using compliant fuel (using the closed loop technology, those systems can provide even cleaner environment that compliant fuel per ton fuel used), why we are of the opinion that the EEDI need to compensate ships that utilizes EGCS. * MEPC70/INF.9, Supplemental Marine Fuel Availability Study, page 101

43 Annex 1, Page 41 Submitter IACS CESA Information **Krantz, Gustav, Transoleum, CO2 and sulphur emissions from the shipping industry CO2 emissions related to the fuel switch in the shipping industry in Northern Europe, Oct 2016, page 1 The following proposals appeared during the discussions and could be considered - however we have no further information at the moment. The two last issues will not have any influence on the EEDI however both can give considerable energy savings on ships. - Trim optimization - although it is an operational issue - hard ware is required to be installed. The use of this hard ware will result in energy savings - Use of Air Control Air Quality Management Systems in accordance with SOLAS Chapter II-2 regulation 20 and MSC/Circ.1/1515 appendix 3 - Use of LED light IACS does not have specific input. However, IACS member would like to bring the fact that the carbon factor (CF) given in the 2014 Guidelines on the method of calculation of the attained EEDI for new ships concerning the calculation method for the EEDI (MEPC.263(68), as amended) does not differentiate the source of fuel (renewable or not) - Categorization/Type of - Short-description <General comment: the description are sometimes far from being short and should be reduced in volume> - ratio* (%) - (New item) ratio when using with other energy saving * **(%) <Comment: This interrelation and goal conflicts between different can be qualitatively described, but are extremely difficult to quantify without clear definition of ship type, transport task, design conditions etc.; should be tackled by sample ship concepts, e.g. as described and assessed in the EU R&D project JOULES (see for the time being the interrelation of combined is only addressed qualitatively> - The year expected for in-use of that technology* - (New item) Current status of application* (normally used / sometimes used/ rarely used/ never used) - (New item) Future prospect of the applications/adoptions of the technology * *** - (New item) Estimated cost for applying the technology* **** (US dollar) <Comment: this item requires additional guidance, e.g. definition of shiptype, size or alternative parameters, such as US$ per installed kw; in addition the intent of this request is unclear, because the assessment of feasibility and availability should be independent of commercial considerations>

44 Annex 1, Page 42 Request 2-3 : If members have any other publicly available and verifiable information related to energy saving technology, please provide these information. Submitter Australia Finland Germany Japan Netherlands Information Australia considers that in general, the energy saving that currently exist may have not yet been fully explored. Therefore, firstly understanding the operating environment of a vessel, and then choosing the technology to suit the situation, is essential. In particular, the possibilities of burning multiple fuels simultaneously (e.g. gas and fuel oils (either bio fuel or residual fuel oils)) may warrant consideration. From our understanding, this is already a common practice in the offshore oil and gas industry. Alternatively, burning gas in port and residual fuel oils whilst at sea on passage, could also be looked at more closely (noting that from our understanding there may already be cruise ships running gas turbines in port for power generation, then running diesel engines directly coupled to alternators for power generation at sea). A similar approach for a bulk carrier or tanker could be to run an alternator on gas in port for power generation then a shaft alternator and turbo alternator utilising waste heat recovery from the main engine exhaust gas at sea. Our view is that it may be a case of understanding the operating profile of the vessel, then choosing machinery that can maximise efficiency under those circumstances when designing a vessel. At the moment we do not have any other publicly available and verifiable information related to energy saving. The European JOULES project provides insights and findings on the effect of innovative and energy saving. The design study PERFECt evaluates the potential reduction in GHG emission by the application of Combined Cycle Gas and Steam Turbine System using the example of a Container Ship: No Comment General remarks: 1. the improvement ratio as presented in the annex 2 of MEPC 68/INF.38 is rather high when compared to the study as presented in MEPC 70/INF.32 (see below). This may be caused by the fact that the data in MEPC 68/INF.38 is not verified. 2. The content of annex 2 of MEPC 68/INF.38 is quite exhaustive. However there is one key element that is missing: it is not showing which of the energy saving can be used simultaneously. Neither is it showing how the improvement ratios should be combined once simultaneously installed. Reference is made to the attached documents: - MEPC 70/INF.32 (the Netherlands), please note that the reduction percentages in this document are based on the EIV. There may be some differences with the calculation of EEDI. More specific for ships with two-stroke engines the reduction percentages are about 10% higher (which means: more energy efficient). This is caused by the specific fuel oil consumption: for the EIV this is set at 190g/kWh. This number will do for medium speed engines. However the bigger bulk

45 Annex 1, Page 43 Submitter Information carriers and containership (equipped with slow speed two-stroke engines) will have have a sfoc between g/kWh. When this sfoc is taken into account the reduction percentages for these bigger ships would be about 10% higher. - E.A. Bouman et al. / Transportation Research Part D 52 (2017) Republic of Korea Russian Federation Sweden United States CESA IACS ICS We have no specific information. The current IMO procedure for EEDI calculation (MEPC.212(63)) does not consider take into account the possible fuel economy of the ship due to use of outboard cooling systems (box coolers and keel coolers) and omission of the cooling pumps. Electricity consumption during the course of that ship is reduced to 20% and the overall fuel consumption of the ship on the move is reduced to 1%. We have no new information at this moment. We have no additional information. Energy saving / CO2 emission reduction have been assessed in the EU R&D project JOULES also addressing fuel related life cycle aspects including the effects of energy conversion and distribution. Public deliverables are available for download on the JOULES website: The IMO Global maritime energy efficiency partnerships (GloMEEP) offers an Energy Efficiency Technologies Information Portal addressing some of the (excluding alternative fuels and energy converters) with mostly comparable reduction potentials: The GloMEEP assessment of the technical maturity is, however, inappropriate because the technical availability and feasibility is confused with the uptake of technology by the industry, which is governed by commercial considerations. IACS does not have specific information. ICS considers that as increasingly complex systems are used to improve efficiency it will be important to ensure that the ships EEDI value is representative of what the ship will achieve in service. ICS considers that Annex 3 2 provides a comprehensive list of available to improve energy efficiency. We do however have some concerns with this list: The real determining factor as to what degree of efficiency these will deliver is not so much about the technology itself but how it is integrated with the other systems on-board and its suitability for the intended operating profile of the ship;

46 Annex 1, Page 44 Submitter Information Some of the listed are of limited application and reported results are variable, in some cases there appears to be limited independent verification for claimed efficiency improvements; There appears to be an implicit assumption that some of the reported efficiency improvements can be used in conjunction with figures found in the EEDI database to extrapolate future EEDI values, we would urge caution before arriving at any such conclusion and would advise that any analysis of the potential value of the listed will require a significant effort and will have to consider limiting factors and shipboard integration. One of challenges we will face in this work is that if considering a particular technology it is often the case that for each example of its application purporting to demonstrate positive performance improvements there is often another example which demonstrates that the same technology displayed no discernible improvement when installed on another ship. Therefore the group could arrive at erroneous conclusions unless the efficacy of each technology is considered in detail, including identifying those applications which are either well suited or unsuitable for a particular technology. This raises a consequential problem for the group in that when considering many of these it is very easy to conflate the EEDI and EEOI. This work is concerned with the EEDI, we need to be mindful to avoid mixing operational and design improvements, however this may be difficult since as systems become increasingly complex it may become difficult to separate operational and design matters. More work is required to develop guidelines for how new and innovative will be included within EEDI calculations, for example battery hybrid propulsion packages. INTERFERRY As noted under question 1-1, introduction of an x% more efficient technology on one ship design does not equate to an acrossthe-board assumption that said technology would yield x% improvement if widely adopted. Hybridization, for instance, could generate significant efficiency gains for a given diesel-electric design, but not obviously so for a diesel mechanic design. Since diesel-electric is inherently less efficient than diesel-mechanic, especially compared to two-stroke engines, it would be incorrect to assert that hybridization would automatically yield a given percentage improvement within a segment. RINA As noted above, RINA thinks that there is a self selection bias inherent in all reports of energy saving because the instances where no savings were found are not reported, although anecdotal evidence would suggest that this is widespread. There are also no robust global standards for measuring energy savings (although ISO is a good start) that give meaningful results for all ship types. Reported savings obtained are sometimes due to ships being operated off design point, and the are correcting for this sub-optimal condition.

47 Annex 1, Page 45 Submitter Information RINA also reiterates that TRL (technology readiness levels) ought to be provided for all candidate, and those that show the most potential and widespread applicability should be the subject of intensive and aggressive research, development and full scale testing that considers not only the efficiency gains, but the safety and reliability of such.

48 Annex 1, Page 46 Request 3 : Members are invited to provide comments on the proposed new ice class correction factors for capacity fi contained in MEPC 71/5/6 and MEPC 71/INF.16. Submitter Australia Finland Germany Japan Comments (with rationale) No comment No comments at this point of time. No comments. Table 3 of MEPC 71/5/6 contains average Cb values for less than 120,000 DWT, but there are ships of over-120,000 DWT cargo ships with ice class. It is necessary to include average Cb value for over-120,000 DWT ships. Japan doesn t have other specific comments on the proposed fi correction factors contained in MEPC71/5/6 at this time, but is of the view that correction factor fi and fj should be considered together. Therefore, Japan would like to reserve its right to make our comment on both fi and fj after having a draft calculation method for fj. Netherlands In general we support the Finnish/Swedish proposal to replace the existing ice class capacity correction factors in the 2014 Guidelines with new factors and proposal to allow the use of the ship specific voluntary structural enhancement correction factor to calculate the ship specific ice class correction factor for capacity. In this way, the ice class capacity correction factor could be determined in a more accurate way for all ships having a Finnish Swedish Ice Class or any other ice class. We do however have 2 comments: 1. It seems so that the Fi factor for deadweight correction is not very accurate for small ships. Ref is made to the example ship of 3000 DWT as listed in the documentation TRAFI. Along the proposed formula for ice 1A: Fi = /DWT, the correction factor would be This means that the assumed deadweight loss (equal to the extra construction weight due to the ice belt steel) would be: (3000 * ) 3000 = tons According to the calculations made for each ship (TRAFI) the steel weight increase for ice 1A is 54 tons. Concluding: it seems so that for smaller ships the formula for Fi leads to overrating the deadweight penalty the small ice class ships have. 2. The methodology for capacity correction due hull lines adaption / displacement loss seems very rough if it is based solely on Cb. Cb is not only governed by bluntness of bowlines (affecting ice breaking). L/B ratio, B/T, transom immersion all have large influence on attained Cb but are hardly affecting icebreaking. Ships with high Cb (0.85) can be created with sharp entrance angles in the bow (favorable for icebreaking),in combination with large L/B. Comparing to a Cb of a reference design seems therefore very rough, and only fitted for a very slender band of ships with average main dimension ratio.

49 Submitter Republic of Korea Russian Federation Sweden United States CESA IACS MEPC 72/INF.12 Annex 1, Page 47 Comments (with rationale) We consider the modified correction factor Fi for ice class vessels proposed in document MEPC 71/5/6 will be more precise and appropriate to apply EEDI for Ice class vessels. No comments at this point of time. No comments at this time. No comments at this time. no comment IACS believes this paper takes an appropriate approach to fix and improve on the present ice class correction factors. Accordingly, this paper can be supported in general (Extract from IACS position for MEPC 71) ICS recommends support for the proposed new ice class capacity correction factors contained within MEPC.71/5/6. ICS INTERFERRY INTERFERRY has no input to provide on this matter RINA RINA has no comment on the subject at this time

50 Annex 1, Page 48 Request 4 : Members are invited to provide comments on the proposal contained in paragraph 4 of document MEPC 71/5/2, especially for a suggestion on the methodology to be used on how to define a margin to the reference lines of ships having an ice class and related background information. Submitter Comments (with rationale) Australia No comment Finland In Annex 3-2 there is a comment sheet for the information of energy saving. Our proposal is that, the possibilities to apply the energy saving listed in this annex to ships having an ice class should be analyzed in detail in order to evaluate, if they can be applied in these ships. This analysis could then be used as a basis to define a margin to the reference lines of ships having an ice class. Germany No comments. Japan There are certain number of ice classed ships with energy saving devices, especially for lower ice classed ships. It is not always impossible for Ice classed ships to introduce and apply energy saving devices. On the other hand, higher ice classed ships, which would be considered more difficult to introduce and apply energy saving devices, have to install larger engines, because of the class requirements and it is not possible for such ships to reduce their engine power. Thus, the effect of energy saving devices could only be reflected to Vref at 75 percent MCR, and this means the effect on the attained EEDI is quite limited. Therefore, careful consideration should be done for introducing additional margin of EEDI requirement for ice classed ships because of the difficulty in applying energy saving devices. In further considering the additional margin for the ice classed ships, consideration and evaluation based on actual effect of the limitation of energy saving devices on ice classed ships should be done. Netherlands We can in general support the proposal to revise the reduction factors for EEDI requirements as per paragraph 4 of MEPC 71/5/2. Republic of No comment Korea Russian See part 1 of RF Annex Federation Sweden We can support the proposal in general, but don t have any suggestion on methodology. United States No comments at this time. CESA no comment IACS 1) When regulations other than the Finnish-Swedish Ice Class Rules are applied, it may be difficult to identify which ice classes are to be considered as "higher than IA super"; and 2) it is suggested that the word cargo be deleted from the proposed amendments to MARPOL Regulation 19.3 (paragraph 1 in the annex to MEPC 71/5/7), since apparently the intention of the Russian Federation was to exclude all ships (including passenger ships) having ice class higher than IA Super from the EEDI calculation (see MEPC 71/5/7, paragraph 6).

51 Submitter ICS MEPC 72/INF.12 Annex 1, Page 49 Comments (with rationale) ICS recommends that the possibilities for applying energy saving listed in annex 3 2 to ships having an ice class should be analyzed in detail to evaluate which, if any, can be applied to such vessels. This analysis could then be used as a basis for defining the margin to the reference lines for ice classed vessels. INTERFERRY INTERFERRY has no input to provide on this matter RINA RINA has no comment on the subject at this time

52 Annex 1, Page 50 Request 5 : Members are invited to provide a concrete proposal on how ice classes higher than IA Super, which could be excluded from the EEDI regulation, should be defined, taking into account MEPC 71/5/2, MEPC 71/5/7 and MEPC 71/INF.7. Example of inclusion/exclusion based on the proposed definition may also be useful for clear understanding of the definition. Submitter Australia Finland Germany Japan Netherlands Republic of Korea Russian Federation Sweden United States Comments (with rationale) Australia acknowledges the difficulties associated with applying EEDI to vessels with ice class above 1A as outlined in MEPC 71/5/2. Australia also notes that there may be a shift toward the International Association of Classification Societies Polar Class notations being more universally adopted for new ships moving forward and queries if this should be taken into consideration in the definition. The current regulations exclude cargo ships, which are designed to break level ice independently with a speed of at least 2 knots when the level ice thickness is 1.0 m or more having ice bending strength of at least 500 kpa, from the EEDI regulations (see regulation 19.3 of MARPOL Annex VI). According to our understanding, these ships are ice-strengthened cargo ships designed to sail independently in medium first-year ice conditions (first-year ice of 70 cm to 120 cm thickness). This means that ships, which meet the category A regulations of the Polar Code could be exempted from the EEDI regulations. According to the Polar Code, category A ship means a ship designed for operation in polar waters in at least medium first-year ice, which may include old ice inclusions. This kind of exemption would mean that ships having at least an ice class PC 5 of IACS or equivalent (see regulations and of the Polar Code), and which otherwise meet the provisions of the Polar Code, would be exempted from the EEDI regulations. This kind of exemption would be easier to verify than the current regulation 19.3 of MARPOL Annex VI. No comments. No comment A distinction could be made between the total installed power (also needed for ice breaking), and the power needed to maintain an economical cruising speed in open water. EEDI determination for these special ships could be based on the power needed for cruising speed, and not on speed at certain of total installed power. It is deemed necessary to make a clear definition for the term higher than IA Super, before discussing whether or not excluding ships with ice classes higher than IA Super from the EEDI regulations. In this regard, the Resolution MEPC.245(66), 2014 guidelines on the method of calculation of the attained Energy Efficiency Design Index(EEDI) for new ships could be amended adding a definition. With the definition, we also may add an application criteria in the same resolution such as; Ships with ice classes higher than IA super are possibly deemed as ships having ice-breaking capability. See part 2 and 3 of RF Annex It should be possible to use the ice classes in the Polar Code that are higher than IA Super that could be excluded from the EEDI requirement. No comments at this time.

53 Annex 1, Page 51 Submitter CESA IACS Comments (with rationale) no comment IACS was unable to discuss this issue in depth owing to the late participation in the CG. The following view was expressed by one member (not yet fully agreed by IACS members) but may provide basis for discussion in the future round of CG. IACS will provides its consolidated comment at a future round. ICS Ice class IA-Super ships have such structure, engine output and other properties that they are normally capable of navigating in difficult ice conditions without the assistance of icebreakers. The design requirement for ice class IA-Super is a minimum speed of 5 knots in the brash ice channels of (thickness at mid channel) HM = 1.0 m and a 0.1 m thick consolidated layer of ice. Requirements for ice class higher than IA-Super could include e.g. a speed of above 5.0 knots in ice channels with HM > 1.0 m and consolidated layer of ice with thickness exceeding 0.1 m. The current regulations exclude cargo ships, which are designed to break level ice independently with a speed of at least 2 knots when the level ice thickness is 1.0 m or more having ice bending strength of at least 500 kpa, from the EEDI regulations (MARPOL VI regulation 19.3). These ships are ice-strengthened cargo ships designed to sail independently in medium first-year ice conditions (first-year ice of 70 cm to 120 cm thickness). In the Polar Code, a category A ship is one which is designed for operation in polar waters in at least medium first-year ice, which may include old ice inclusions. If looking at the already noted exclusion for ice class ships in regulation MARPOL VI regulation and also the definition of a category A ship in the polar code then ICS considers that ships meeting the category A regulations of the Polar Code could be exempted from the EEDI regulations. Therefore, we would recommend that ships having at least an ice class PC 5 of IACS or equivalent (see regulations and of the Polar Code), and which otherwise meet the provisions of the Polar Code should be exempted from the EEDI regulations. INTERFERRY INTERFERRY has no input to provide on this matter RINA RINA has no comment on the subject at this time ***

54

55 Annex 2, Page 1 ANNEX 2 COMMENTS SUBMITTED FROM PARTICIPANTS FOR ROUND 2 Request 2 : Based on the preliminary assessment given by members, members are invited to provide comments on the following specific points: Request 2-1 : There are several opinion that bulkers and tankers, especially for larger bulkers and tankers, may have difficulty of meeting Phase 3 requirement. Members are requested to provide their views about possible reason of this and a necessity of special treatment such as correction factor or different reduction rate for Phase 3 requirement. Comments for each typical ship sizes such as Handy bulker, Handymax bulker, Panamax bulker, Cape-size bulker, VLOC, MR tanker, Panamax tanker, Aframax tanker, Suezmax tanker and VLCC are highly recommended. Submitter Australia Comments (with rationale) Based on current data, it does appear that there are a certain proportion of Tanker, Containerships, General Cargo, Bulk Carrier and Gas Carriers in the database provided which meet or exceed the Phase 3 reference line currently. On this basis, it is possible that with improvements in the ability to meet or exceed the reference lines will also improve.

56 Annex 2, Page 2 Submitter Denmark Comments (with rationale) In line with several delegations Denmark recognizes the difficulties to meet the EEDI phase 3 requirements for large bulk carriers and tankers. In the following we give our opinion on this and suggest a special treatment for these ships. The relation between DWT and EEDI on bulk carriers keel-laid are illustrated in figure 1. The scale is logarithmic and EEDI reference lines appear as straight lines. The data are separated in 3 categories: DWT <20.000, < DWT < and DWT > When calculating the tendency lines of the ships in each group, it is observed that the slope of the line for ships above DWT (red line) is smaller than the slope for the group between and DWT (blue line). The same observation is found when comparing tankers keel-laid , as shown in figure 2. The slope of the line for ships above DWT is also in this case smaller than the slope for ships between and DWT. These observations illustrate the problems related to EEDI reduction on large bulk carriers and tankers. The large ships will, as a rule, find it much more difficult to reduce the EEDI compared to smaller ships. The reason for this is likely related to the power required to maintain good overall performance of the ship. The dominant design criteria on ships are the DWT and the speed. However, on very large ships with deadweight of tons or more, the power required for proper acceleration or deceleration may be even more important to maintain good maneuvering performance. A small ship has a relative large power compared to the ship size and the ship will be easy to accelerate. The opposite situation is found for large ships as they have a relative small power compared to the ship size. In this case the lack of power makes the acceleration difficult and may be severe in situations, which are not unusual for the ship. Although the maneuverability of ships is discussed in relation to the minimum propulsion power of ships, this should also be included in the regulation for reduction on EEDI. It should be kept in mind that large ships have very good energy performance and it would be appropriate to keep a realistic and safe performance requirement for these ships. A suggestion to handle the requirements for large bulk carriers and tankers could be to introduce the same kind of relaxations given to ships between and DWT. In figure 3 and 4 a relaxation is shown in the interval between DWT and DWT. At DWT the EEDI reduction is maintained for phase 1, 2 and 3 but at DWT the EEDI of phase 1, 2 and 3 will keep the EEDI requirement for phase 0, i.e. the EEDI reference line. A linear interpolation is suggested between these limits. Other ship types, as container ships, may meet the same problems for large ship size, but these ships have a traditionally high power and may not meet the problems in the size they are designed for.

57 Annex 2, Page 3 Submitter Comments (with rationale) Figure 1

58 Annex 2, Page 4 Submitter Comments (with rationale) Figure 2

59 Annex 2, Page 5 Submitter Comments (with rationale) Figure 3

60 Annex 2, Page 6 Submitter Comments (with rationale) Figure 4 Germany Pure statistical analyses of the publicly available data indicate that larger Bulk Carriers and Tankers tend to face higher challenges to comply with Phase 3 requirements than smaller ships of these types. This perception needs also to reflect the initial situation for the development of the reference lines, hence the status of ship building and market at that time. While there was a tendency for hydrodynamic optimization for larger ships (more expensive / higher fuel lifecycle costs) already before introduction of the EEDI, higher efforts were seen after the implementation of the EEDI for smaller Bulkers and Tankers. This would party explain the current situation where the optimization margin for larger Bulkers and Tankers is already widely consumed while smaller ships even exceed Phase 3 requirements. Also, for smaller Bulker and Tanker designs, a reduction of design speed can be observed which was not the case for large ships of these types.

61 Annex 2, Page 7 Submitter Comments (with rationale) Against this background, one possible option is to redraw the EEDI reference line representing a certain share of the best performers of phase 0 (here the most representative number of certified EEDI values is available) and redefine this as base reference line. Using a typical design speed DWT curve for Bulkers and Tankers, the Froude number has a decreasing characteristic for larger ships resulting in a higher influence of the frictional resistance to the total resistance. Frictional resistance cannot be reduced by hydrodynamic optimization of the ship, reducing the choice of technical options for the reduction of the EEDI. Additionally, the minimum required propulsion power needs to be considered especially for the larger ships of these types as reflected in the following graph. The graph shows the minimum reference speed to be reached by a ship of a certain size to fulfill the minimum required propulsion power and the EEDI requirement (note that using the minimum required power line instead of the simplified assessment, the graph tends to be slightly conservative, i.e. EEDI reference speed would be slightly less for larger ships).

62 Annex 2, Page 8 Submitter Comments (with rationale) The graph shows the following curves: MCR_ME_MIN: Minimum required propulsion power according to MEPC.232(65) (Minimum power lines) V_KRISTENSEN: Average design speed of bulker fleet according to Technical University of Denmark 2012 (Hans Otto Kristensen) PHASE 0-3: EEDI reference speed to be achieved by a Bulker in order to fulfil the required EEDI while fulfilling the minimum required propulsion power (according to minimum power lines) In contrast, typical design speed DWT curve for Bulker/Tanker for the time before implementation of the EEDI is shown in the following figure:

63 Annex 2, Page 9 Submitter Comments (with rationale) Source: Technical University of Denmark 2012 / Hans Otto Kristensen It can be seen that the stricter EEDI requirements in combination with the minimum required propulsion power would force these ship types in phase 3 to sail / design them for 4 to 5 knots faster compared to the time before the EEDI was implemented. Consequently, there is a conflict between these two requirements which apparently cannot be solely solved by the hydrodynamic optimization of the ships. This physical conflict could either be solved by a larger technological step towards the application of alternative fuels (e.g. LNG)

64 Annex 2, Page 10 Submitter Japan Comments (with rationale) / propulsion systems (e.g. wind being most suitable for slower ships) or the implementation of a shaft power limitation. The latter would decouple the minimum required propulsion power (only available in emergency situations) from the EEDI power. For bulkers and tankers, there was limited number of data for Capesize bulkers, Suezmax tankers or larger ships, in developing the reference line. Lack of the data has led to the stringent reference lines for the larger ships, which are higher than the average of attained EEDI values of these larger ships. Figure 2 and 4 show distribution of EEDI value compared with reference line for bulkers and tankers. It can be found that about 68 percent of bulkers over 120,000 DWT and tankers over 200,000 DWT are above the reference lines. Furthermore, new ship size called VLOC, of which deadweight is larger than 400,000 DWT, has newly come out after 2008 but which had not been taken into account when developing the reference line (Fig 1, 5 and 6). In general, 30 percent EEDI reduction rate is quite high requirement for bulkers and tankers, but this makes more severe for large bulkers and tankers to archive 30 percent reduction of attained EEDI. Thus, it is difficult to apply the same requirement for all ship sizes of bulkers and tankers, and the larger ships should be regarded as different categories when considering the requirement of attained EEDI for phase 3. Further, we should take into account the minimum power requirement because bulkers and tankers already have slow design speed so that there is small margin of reducing MCR and improving attained EEDI. Thus, it should be noted that unnecessarily stringent minimum power requirement will be a barrier to GHG emission reduction from new ships.

65 Ratio of Ships EEDI Values MEPC 72/INF.12 Annex 2, Page 11 Submitter Comments (with rationale) No data for >400,000DWT , , , , , , ,000 DWT Reference line Phase 0 Phase 1 Phase 2 Phase 3 DOD data Fig. 1 EEDI value used for developing the reference line (bulkers) 100% 30%~40% 80% 20%~30% 60% 40% 20% 0% 10%~20% 0%~10% -10%~0% -20%~-10% -30%~-20% -40%~-30% -50%~-40% DWT ±0% Line Fig. 2 Distribution of EEDI value compared with reference line (bulkers)

66 Ratio of Ships EEDI Values MEPC 72/INF.12 Annex 2, Page 12 Submitter Comments (with rationale) , , , , , , ,000 DWT Reference line Phase 0 Phase 1 Phase 2 Phase 3 Data "Reference Line" Fig. 3 EEDI value used for developing the reference line (tankers) 100% 30%~40% 80% 20%~30% 60% 40% 20% 0% 10%~20% 0%~10% -10%~0% -20%~-10% -30%~-20% -40%~-30% -50%~-40% DWT ±0% Line Fig. 4 Distribution of EEDI value compared with reference line (tankers)

67 DWT MEPC 72/INF.12 Annex 2, Page 13 Submitter Comments (with rationale) 500, , , , ,000 0 y = x R² = , , , , ,000 DWT DOD data Fig. 5 DWT of bulkers built from 1999 to 2008

68 DWT MEPC 72/INF.12 Annex 2, Page 14 Submitter Comments (with rationale) 500, , , , ,000 0 y = x R² = , , , , ,000 DWT DOD data Fig. 6 DWT of bulkers built from 2009 to 2016 Netherlands We are not of the view that bulkers and tankers have difficulty in obtaining phase 3 requirements. We would like to draw the attention to table 6-1 of MEPC 70/INF.32. In the study the table was taken from, the EIV was used instead of EEDI. However it was concluded there is no big difference between the two and therefore in our view results are also valid for the EEDI. The research looked into increased efficiency from better hull and propeller design and the use of Energy Saving Devices. Although the number of ships in the study were limited in our view these results are actually supported by the data provided for existing ships since it is clear from the data that there is room for improving hull and propeller designs (and perhaps even better improving the design of the installed power train) and furthermore that the use of ESDs is low for these types of vessels.

69 Annex 2, Page 15 Submitter Comments (with rationale) Norway There was an argument brought forward that the Minimum Power Guidelines may be a reason for difficulties in obtaining the reduction rates. However in our view these guidelines need more refinement in that sense that it is not the installed propulsion power that is the overriding factor for safe maneuvering of ships in adverse weather conditions but it is the power the propeller can deliver at very low speeds for keeping the ship maneuverable. We refer to MEPC 70/ INF.28. We do agree however that the minimum power the propeller should deliver at low speeds for the ship to remain maneuverable has to be checked against the EEDI, so that the (maximum) power that can be installed in compliance with EEDI does not drop below the power to be installed in accordance with (refined) MPR guidelines. There have been opinions that phase 3 could be difficult to achieve for tank and bulk vessels. This concern is believed to be caused by the need to satisfy both the EEDI and the minimum propulsion power requirement. For the minimum propulsion power requirements, a provision was added to regulation 21 in chapter 4 of MARPOL Annex VI, stating that: For each ship to which this regulation applies, the installed propulsion power shall not be less than the propulsion power needed to maintain the manoeuvrability of the ship under adverse conditions [ ]. For bulk carriers and tankers a guideline on how to compute the minimum propulsion power is developed [REF]. This result in a more stringent interpretation of the minimum propulsion power for these vessels types than for others. The easiest way of improving EEDI for most vessels is to reduce the installed power, however current designs are close to the minimum propulsion power requirement and are not satisfying the EEDI phase 3 requirements. This means that complying with EEDI is not possible by simply reducing the installed power. Assuming a recent, but traditional design is close to the minimum propulsion power limit there are three main options for improving the EEDI: 1. Improving hull and propulsion design to achieve higher speed using the same engine power 2. Alternative fuels 3. Improve performance in harsh weather to reduce minimum propulsion power requirement and hence reduce installed power It should be noted that efficiency improvement taken out as speed increase only has about 1/3 effect on the EEDI as if the gain was taken out as reduced installed power. The following options, related to hull and propulsion design (i.e. the first point (1) in the list above), can be used to improve EEDI: Alternative fuels (can have a large effect on the EEDI) Reduce installed power Hull and propeller optimization Improved machinery efficiency through design or operation optimization Waste heat recovery

70 Annex 2, Page 16 Submitter Republic of Korea Comments (with rationale) PTO/PTI - If you use a combined PTO/ PTI, the normal operating mode shall be used in EEDI. This means you can install a large PTI for use in bad weather or to catch up delays without being penalized in EEDI. When PTO/PTI is installed, normally the PTO mode is applied in EEDI calculations because it is the most efficient mode and it more closely resembles normal sea going condition, which is what EEDI is supposed to be used for. Air bubble lubrication does not affect the limit for minimum propulsion. However, it will have a positive effect on the EEDI on sea trial. Possibility for further technology development If minimum propulsion power/manoeuvrability is an issue the examples below can allow bulk carriers and tankers to reduce the required minimum propulsion power and hence meet the EEDI requirements by reducing the installed power. The following list of measures thus complements the third point (3) in the list of options for reducing the EEDI on the previous page: CPP will increase thrust at low speeds (able to take out full power) - FPPs as opposed to CPPs can usually not take out full power at low speeds due to torque limitations. - CPPs can produce up to 50% more thrust in bollard pull than FPPs. Ducts will increase produced thrust at low speeds, but reduce thrust at high speeds High lift rudders e.g. flap rudders, fish tail rudders or similar will increase steering force - High lift rudders e.g. flap rudders generate up to 50% more steering force than conventional rudders. Azimuth propulsion can produce full thrust in any direction. Reduce added resistance in a seaway. In addition, ice correction factors can be relevant for bulk carriers and tankers if built within the Common Structural Rules (CSR) regulations. To summarize: Without a detailed study, it is believed that the phase 3 requirements can be met without introducing a correction factor for tankers and bulk carriers. In addition to alternate fuels, improving the performance in harsh weather and hence reducing the minimum propulsion power requirement is a promising option. Regarding Bulk Carriers and Tankers, especially for VLCC and Suezmax Tanker, there are many cases that available energy saving which are proven in the market have been already applied to reduce attained EEDI for these kinds of vessels, and it might be difficult to optimize power of vessels to improve energy efficiency because of the minimum propulsion power requirements. Furthermore, the range of principal particulars including main dimensions and speed have been developed for long time and formed within standardised range, thus significant change of design may not be feasible. At the moment, in the current framework of regulation, the only way to comply the requirements of EEDI Phase 3 is to use LNG fuel, but LNG Fuel Supply increases CAPEX and the infrastructure, rules and guidelines for LNG bunkering have not been developed enough for service on time.

71 Annex 2, Page 17 Submitter United Kingdom Comments (with rationale) We also need to carefully consider reduction rate of bulk carriers such as handy, handy-max and Panamax bulkers with minimum power requirements which is under development since many bulk carriers are still in range of 10 to 20 percent reduction of attained EEDI. For these reasons, it seems to be difficult to meet the requirements of EEDI Phase 3 for Bulk Carriers and Tankers, and it to be needed for an in-depth review of the adjustment of the reduction factor for EEDI Phase 3 for all sizes of these vessels. The EEDI reference line set for bulk carriers and oil tankers are based on the usual service speed which is between 10 to 15 knots. Hence, these types of ships might have difficulties to meet phase 3 Required EEDI s requirement due to there is lack of room for improvement by deploying slower design speed ships unlike other ships such as container ships-. This may be the possible reason. There are EEDI data base indicating bulk carriers and oil tankers of certain sizes are able to meet phase 3 requirement even without deploying innovative on hull and machineries design which is encouraging news. Nevertheless, the larger bulk carrier of 150,000 DWT and oil tankers of 250,000 DWT, even the best performance hull and machineries design, may not yet meeting phase 3 requirements. Based on above, it is reasonable and sensible to consider a two tiers approach for phasing in phase 3 requirement. Larger bulk carriers (above 150,000DWT) and larger oil tankers (above 250,000) to meet the phase 3 requirements beginning from 2025 and, advance the phase 3 requirement for other sizes bulk carrier and oil tankers to the beginning of 1 st Jan We also like to bring in Safe return to port principle. This principle should be taken into consideration during this discussion by including minimum power. United States USG: The data from the EEDI database show there are ship designs that currently do not meet the Phase 3 reduction rates at this time. However, the requirements apply only for new ships beginning in Because many ship designs already meet the requirements, in all ship size classes, it is reasonable to consider the reduction factors to be achievable overall. We think the data support retaining the 30% reduction factor. It may even be possible to advance the effective date to 2022 or 2023, given the number of ship designs that already achieve the requirements. EC The new version of the EEDI database confirms the view the European Commission expressed in Round 1 that in general tankers and bulkers can meet the phase 3 requirements. The statistical analyses show that for both ship types, the best-performing designs (expressed as the average performance of the best 10% compared to baseline) are in case of tankers already today better than the phase 3 requirement (35% below reference line) respective very close to this requirement in case of bulk carriers (27% below reference line), both without significant use of innovative. However, more detailed analyses of size categories indicate that the best-performers within the largest categories of tankers and bulk carriers are not yet meeting the phase 3 requirements. This is in particular the case for bulk carriers above DWT and tankers

72 Annex 2, Page 18 Submitter Comments (with rationale) above DWT. The other size categories indicate that best-performers are already meeting or close to meeting phase 3 requirements, even without use of innovative. In case the advancement of phase 3 to 1 January 2022 is confirmed, the European Commission could accept excepting these two categories from the advancement to Thus, for bulk carriers above DWT and tankers above DWT the phase 3 requirements would only apply as from Ship type/ size category Bulk carrier DWT Bulk carrier > DWT Tanker DWT Tanker > DWT Tanker > DWT Number phase 0 & 1 Average performance 'Early compliance' Use of innovative ships (mandatory of 10% best (in % phase 3 EEDI) below ref line) % 1% 0% % 0% 0% 86 33% 16% 0% 40 26% 0% 0% 66 21% 0% 1% CSC The EEDI scores in the latest release of the IMO EEDI data base corroborate earlier findings by the CSC of significant and in some cases massive early over-compliance with Phase III requirements. See Table 1 below, which shows that not only are many ships already complying with Phase III, but the performance of the 10% best ships outpaces the Phase III requirement by 27% for containership, 4% for tankers, 11% for gas carriers and 27% for general cargo ships. And by, on average, 10 years ahead of time.

73 Annex 2, Page 19 Submitter Comments (with rationale) Table 1 Analysis of EEDI scores of recently built ships Ship Types analysis from 2011 to 2017, inclusive (mandatory phases) Bulk Carriers Container Ships Tankers Gas carriers General carg ships Total number of ships in mandatory phase Distance to EEDI reference line Mean 20% 40% 25% 25% 48% median 20% 43% 27% 24% 50% Share of ships with EEDI 30% under reference line 0.3% 69% 24% 12% 67% Share of ships with 4th term innovative technology 0% 8.9% 0.2% 0% 0% Share of ships with 5th term innovative technology 0% 0% 0% 0% 0% Average distance to EEDI reference line of top 10% best ships Source: Transport & Environment (Clean Shipping Coalition) 27% 57% 34% 41% 57% Clearly, there are variations in the level of achieved efficiencies among ship types and sizes, but what is crystal clear is that there has been virtually no reported uptake of innovative ; for example, just one tanker (out of 622 in the data base) and no bulkers have reported the use of 4 th /5 th term. And yet a comprehensive list of possible was circulated to CG members over 3 years ago in October 2014 (MEPC 68/INF.38) at the beginning of the review process. The significance of this when considering the appropriateness of future requirements can be illustrated by looking at recent analysis of innovative wind like rotors, wingsails, towing kites and wind turbines. (MEPC 68/INF.38 took the view that sail-assisted power does seem to be an interesting opportunity for saving fuel in the medium- and long-term picture. Using real-world AIS data and known weather patterns, one recent study (see Tables 2 and 3) modelled the average energy saving potential of these available, and found that energy savings of up to 9% and 18% at higher speed, and 13% and 24% at lower speeds are possible for large (90,000 DWT) tankers and bulk carriers respectively. Use of these would, in many cases, have improved the EEDI of tankers and bulkers beyond phase III requirements.

74 Annex 2, Page 20 Submitter Comments (with rationale) Table 2: Relative energy average savings across the AIS-recorded voyage profiles higher speed Rotor Wingsail Towing kite Wind turbine Large bulk carrier (90,000 dwt) 17% 18% 5% 2% Small bulk carrier (7,200 dwt) 5% 5% 9% 1% Large tanker (90,000 dwt) 9% 9% 3% 1% Small tanker (5,400 dwt) 5% 5% 9% 1% Source: CE Delft, Table 3: Relative energy average savings across the AIS-recorded voyage profiles lower speed Rotor Wingsail Towing kite Wind turbine Large bulk carrier (90,000 dwt) 23% 24% 9% 4% Small bulk carrier (7,200 dwt) 7% 7% 14% 2% Large tanker (90,000 dwt) 13% 13% 4% 2% Small tanker (5,400 dwt) 7% 8% 15% 2% Source: CE Delft, There appears to be a view in some quarters that the EEDI requirements should be set on the basis of what is known to be possible at the time the requirement is set. CSC fundamentally disagrees with this approach, and believes it runs contrary to the widely accepted purpose of vehicle standards - to drive innovation. That said, it is clear from the above analysis that what is known to be possible at this point in time allows for the bringing forward of the Phase III requirement for all ship types. Taking into account the future emergence of new designs and the potential for the active uptake of innovative a strengthening of Phase III will ensure that the EEDI provides a real incentive to improve efficiency. IACS The correspondence group should recommend to MEPC that the Phase III requirements for all major ship type and size categories (analysed in Table 1 above) are strengthened by bringing the compliance date forward from 2025 to a. In general, these two ship types (bulk carriers and oil tankers) are all difficult to meet phase 3 requirements. Cape size bulkers and larger bulkers(including 400k VLOC )/ Suezmax tankers and larger oil tankers may not meet 30 percent reduction of attained EEDI even if both applying additional energy saving /devices and reasonably reducing design speed of ships.

75 Annex 2, Page 21 Submitter ICS Comments (with rationale) b. New construction VLOC vessels (400k range) significantly exceed the maximum DWT value used by the IMO for the development of the bulk carrier EEDI reference lines. The largest DWT size bulk carrier included in the IMO regression curve has been confirmed at 327k. MEPC.71 EEDI Database (Aug 2017) lists a single 301k BC vessel that complies only with the Phase 1 regulatory limit. No larger capacity bulk carriers verified for EEDI, have been reported to the IMO yet. Comparison of a VLOC s (400k range) attained EEDI value against extrapolation of the current Bulk Carrier EEDI Reference Lines, requires further consideration. c. Special treatment such as correction factor or different reduction rate for large bulker or oil tanker might be a solution which could be further considered. 1. ICS would reiterate that the question is not whether ships such as bulk carriers and tankers can achieve EEDI phase 3, the question is rather how to ensure they can achieve EEDI phase 3 without compromising safe and reliable operation as a result of having insufficient power to maintain manoeuvrability in adverse weather conditions. This is a fundamental point, ICS accepts that if power is reduced without regard for maintaining minimum safe power then these ships could achieve EEDI phase 3. Our position is that to do this would impose unacceptable safety risks and would potentially result in ships and their crews being lost unless concerns about minimum power are addressed. 2. Bulk carriers and tankers have traditionally tended to operate at low speeds, typically knots, unlike some other ship categories which have operated at higher speeds. Before slow steaming took effect, large ocean going container ships typically cruised at 23 27knots. The EEDI reference lines were established before slow steaming took effect with the result that these ships had in effect a significant safe power margin built into their EEDI reference line which facilitates achieving significant improvements in their EEDI values by slowing down and downsizing their engines substantially (large container ships in some cases reduced power by 60% by adopting slow steaming) without risking safety of ships through a loss of manoeuvrability and ship control in adverse weather conditions. The reference lines for bulk carriers and tankers on the other hand were set at levels which did not effectively have the same level of safe power margin built in to facilitate power and speed reductions and improve EEDI values without risking ship safety. Reducing the speed of a slow speed ship will not match the efficiency improvements achieved by high speed vessels because of the relative influences of waves and wind resistance and the fact that slow ships are already operating lower down the propeller power/speed curve. There is evidence to indicate that speed reaches a point below which there is no efficiency gain once other factors are accounted for and there is even a risk of deteriorated efficiency (see 2 3). 3. The example of large container ships is still frequently used to argue that the EEDI phase 3 target is a soft target which ships are already achieving, ignoring the challenges faced by other ship types.

76 Annex 2, Page 22 Submitter Comments (with rationale) 4. ICS cannot analyze EEDI performance in order to propose amendments to EEDI reference lines within the time available for this round of comments, however in principle we would support the consideration of either correction factors for those ships which will struggle to meet EEDI phase 3, or an amendment to their EEDI reference lines. This would allow EEDI reduction rates to better reflect the different challenges faced by different ship types and different ship sizes within these types in order to improve efficiency. Development of such correction factors or adjustments to reference lines would require detailed analytical work, with this work being subject to a review process. ICS would remind the group that a similar amendment was proposed by Greece at MEPC 63, see MEPC 63/4/14, at that time it was not supported based on academic opinion that hydrodynamic improvements could deliver the necessary efficiency improvements however practical experience indicates that the concerns expressed in MEPC 63/4/14 were well found. INTERFERRY INTERFERRY cannot provide any comments on the issues relating to tankers and bulkers having difficulties with Phase 3. INTERTANKO INTERTANKO has collated the data made available from the IMO data base and structured by DWT sizes. The data is only on tankers that have given both mandatory required and attained EEDI values. See table below. One immediate observation is the very small number of tankers delivered with EEDI certificated values for Phase 1. We will need more data for a solid assessment. Another observation is the lack of clarity of the meaning of the reference line value on the IMO database. Is the reference line value the base line of the required EEDI Phase 0) only? We suspect the percentage of the attained EEDI value for ships certified for Phase 1 is still related to the Phase 0 required value. One reason is the consistent average percentage for each group size, be it Phase 0 or Phase 1 certified EEDI attained values. Consequently, Suezmaxes for Phase 1 are only 11.6% below the Phase 1 required EEDI value. Similarly with VLCCs, have an average of 6% below the Phase 1 required EEDI value. INTERTANKO believes that VLCCs may have problems to meet Phase 3 and possible also Phase 2. Based on the data available so far, one could also question whether Suezmax tankers would have difficulties to reach Phase 3 required limit. The reason for the statement above is due to the fact that all improvements achieved for tankers are mostly obtained by de-rating the main engines and of a lesser degree from a small average decrease of the block coefficient. There is little to gain from further reduction of the installed power, particularly with regard to retain a safe minimum installed power (see our comments on 2.3). The consistent average % values shown in the table below also endorse this conclusion. Tanker DWT X 1,000 # of tankers Phase 0 Attained EEDI reduction rate relative to reference line value (Phase 0 certified tankers) (average) # of tankers Phase % 0 - Attained EEDI reduction rate relative to reference line value (Phase 1 certified tankers) (average)

77 Annex 2, Page 23 Submitter Comments (with rationale) % % % % % % % % Some comments for gas tankers. Very limited number (only 4) for Phase 1 gas tankers with large variety in DWT. Not sufficient data to make a good analysis. The small number of data for small and large sizes may explain the large variations on the attained vs required EEDI values currently in the IMO database (see both table and graph below) Gas Tanker DWT X 1,000 # of tankers Phase 0 Attained EEDI reduction rate relative to reference line value (Phase 0 certified gas tankers) (average) # of tankers Phase % % % % % % - - Attained EEDI reduction rate relative to reference line value (Phase 1 certified gas tankers) (average)

78 11,000 15,500 15,500 17,000 17,000 21,000 21,000 21,000 27,500 28,500 30,000 49,500 52,000 54,000 54,500 54,500 54,500 54,500 55,000 55,000 55,000 55,000 55,000 55,000 55,000 55,000 55,000 55,000 55,000 55,000 59,000 59,500 95,000 96,500 99,000 MEPC 72/INF.12 Annex 2, Page 24 Submitter Comments (with rationale) Attained EEDI vs the Required EEDI (%) - Gas tankers- Phase % 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% RINA There is a general regulatory problem governing EEDI for bulk carriers and tankers with regards to minimum power. Resolution MEPC.232(65) had minimum power lines set at a power such that around 95% of the world fleet would meet these minimum power requirements. These lines applied only to Phase 0. These have been revised such that the minimum power requirements were increased in MEPC.1/Circ.850/Rev.1, meaning that now approx. only 85% of the fleet would meet the requirements. This revision is dated 15 July 2015 with a 6 month phase in period it now applies to both Phase 0 and Phase 1 ships. RINA notes that the effect of these increased requirements would be mainly seen in the newer eco-ships, which have in general been designed with lower design speeds than their immediate predecessors.

79 Annex 2, Page 25 Submitter Comments (with rationale) We could illustrate this with an example. A 109,999 dwt tanker with an installed power of kw and Vref of knots would meet the Resolution MEPC.232(65), but not MEPC.1/Circ.850/Rev.1. We could then make some assumptions that if the installed power was increased to match the minimum power required by MEPC.1/Circ.850/Rev.1, we would have a new Pme, and we could estimate a new Vref. Normally we assume that speed is related to power by the cube law, but in a situation where an existing hull form optimized for a certain speed is operated at a higher speed, this relationship may change to one with a higher power law. We have selected a quartic relationship (to the power of 4) to illustrate this. DWT Circ.850 min power (kw) Phase 0 Required EEDI 4.22 Circ.850 Rev. 1 min power (kw) Phase 1 Required EEDI 3.80 Phase 2 Required EEDI 3.38 New Pme Circ.850 Rev.1 (kw) 9849 Phase 3 Required EEDI 2.95 Pae (kw) 496 Attained EEDI 3.08 Original Pme as % of min power 65% % Reduction 27.01% New Vref (knots) cubic New Vref (knots) quartic Installed Power (kw) Pme (kw) 8513 New Attained EEDI (cubic) 3.40 Vref New Attained EEDI (quartic) 3.44 The effect of this change is that a ship design that would have formerly met Phase 2 would no longer do so. However in efficiency terms, as long the ship were operated at the original Vref, there would be no difference in real world efficiency (assuming the SFOC curve of the engine was reasonably flat in the range). The situation with bulk carriers is slightly different because the increase in minimum power requirement has an inflexion point such that smaller ships are affected, while larger ships have a reduced requirement for minimum power. However the position of the inflexion point and the relative gradients of the lines mean that in practice most bulk carriers will have an increased minimum power requirement. There is slight mislabelling in the chart below MEPC.1/Circ.850 should read Resolution MEPC.232(65)

80 Installed Power (KW) MEPC 72/INF.12 Annex 2, Page 26 Submitter Comments (with rationale) Circ.850 Rev.1 Circ DWT (tonnes) The further complication is that RINA understands that the simplified assessment method being proposed by the SHOPERA project leads to even larger values of minimum power, particularly for the smaller bulk carriers. Therefore it seems that on the one hand we are seeking to tighten the EEDI requirement for a number of ship types, while at the same time we are requiring tankers and bulkers to install more power which will lead to a worsening of their attained EEDI.

81 Annex 2, Page 27 Submitter Comments (with rationale) Given the 6 month phase in period, MEPC.1/Circ.850/Rev.1 would appear to have taken effect only for ships ordered on or after 15 January 2016, meaning that very few (if any) tankers and bulkers in the IMO EEDI database reflect this change to the requirements. As long as the data reflecting the current regulations is missing, and minimum power requirements for Phase 2 and beyond are not finalized, we do not have a concrete basis on which to make suggestions for changes to the EEDI implementation schedule or future EEDI phases for bulk carriers and tankers. Regarding a general reason for the difficulty of meeting phase 3 for larger bulkers and tankers, it was known already in that the larger vessels tended to lie closer to the reference line than the smaller ones. RINA would postulate that this partly a function of curve fitting and statistics there are many more smaller ships than there are large ones, so the shape of the curve fit in general would be biased towards the smaller ships.

82 Annex 2, Page 28 Request 2-2 : There are some opinions that smaller container ships may have difficulty of meeting Phase 3 requirement. Members are requested to provide their comment on a reason of this and a necessity of additional measures. Comments with specific threshold value of DWT, such as 35,000DWT or 40,000DWT, is highly recommended. Submitter Australia Denmark Germany Japan Comments (with rationale) Based on current data, it does appear that some smaller containerships have not met the reference line for Phase 3. However, given that many have met or exceed the Phase 2 reference line and a number have exceeded the Phase 3 requirements, Australia considers that the regulation may be achievable given its entry in to force date of In accordance with our experiences it is possible to build container ships of DWT that fulfills the phase 3 requirement. We have no experience with smaller container ships however computer models shows that a speed reduction of around 0.5 knots can make it possible for most of the smaller ships to fulfil phase 3. Statistical analysis shows that a major share of container ships below DWT already exceed Phase 3 requirements. Taking the correct evaluation of the EEDI for granted, it is a proof that for container ships the compliance with Phase 3 is no major burden. This can be explained by the higher design speeds container ships tend to have before the implementation of the EEDI building the statistical basis for the reference curve. The wider spread for smaller containerships and the higher share of ships not over complying with phase 3 can most probably be explained by diverging commercial/operational strategies and a limited incentive to build highly efficient ships. For container ships, it seems that they can meet Phase 3 requirement. On the other hand, careful consideration should be given for strengthening the requirement level or introducing Phase 4 requirement. For large container ships, main engines have become smaller and consequently design speed has become slower due to the economic situation, which lead to the large reduction of attained EEDI value. On the other hand, for smaller container ships, design speed is relatively slow and have not changed (Fig. 1, 2, 3 and 4). In addition, there is few room for these small container ships to install smaller engines, due to their smaller engine power and slower design speed. Thus, it is difficult for the small container ships to achieve the same reduction rate as larger container ships. Therefore, it is necessary to consider different requirement for difference ship size. Threshold deadweight value for larger container ships and smaller containers ships should be considered together with the required EEDI reduction rate based on the trend of design speed and attained EEDI value.

83 MCR [kw] MCR [kw] MEPC 72/INF.12 Annex 2, Page 29 Submitter 120, ,000 Comments (with rationale) 80,000 60,000 40,000 20,000 0 y = x R² = , , , , ,000 DWT Fig. 1 MCR distribution of container ships built from 1999 to , ,000 DOD data 80,000 60,000 40,000 20,000 0 y = 3.121x R² = , , , , ,000 DWT DOD data Fig. 2 MCR distribution of container ships built from 2009 to 2016

84 Speed [knt] MCR [kw] MEPC 72/INF.12 Annex 2, Page 30 Submitter 80,000 Comments (with rationale) 70,000 60,000 50,000 40,000 30,000 20,000 10, FY #N/A Fig. 3 Average MCR of container ships by each ship size and built year FY #N/A Fig. 4 Average speed of container ships by each ship size and built year Netherlands Similar to the reasoning for tankers and bulk carriers we refer to table 6-2 of MEPC 70/ INF.32.

85 Annex 2, Page 31 Submitter Comments (with rationale) Norway Republic of Korea CV-3 is a very small ship of only dwt. CV-2 is moderate with dwt. CV-3 the largest with dwt. While it is true that the reduction rates reduce with the size of the ship in our view the reduction rates as defined now are feasible and we should consider increasing the rates also for smaller container ships. We do consider a distinction between smaller and larger container ships worthwhile. We have not concluded yet on a threshold value but dwt may be rather low. By building a vessel with LNG as fuel the EEDI can be reduced by approximately 25 %. Today 118 vessels run on LNG, 121 projects are on order. There are 63 ports where LNG can be bunkered world-wide with 26 decided to be build (ref. LNGi DNV GL). To assess whether all smaller container ships can run on LNG a feasibility study must be conducted world-wide to assess the cost of installing the infrastructure and which barriers need to be overcome to build the LNG infrastructure. By lowering the speed, smaller container ships that do not meet the phase 3 requirements today will be able to meet the EEDI requirement. However, building small container vessels that sail with lower speeds might affect the competition against trailers, especially for perishable cargo that requires quick transport and delivery. A shift from sea to road transport would lead to higher carbon dioxide emissions, not benefiting the climate or the environment. See also Request 2-3. Regarding small container ships, we have done a case study with two type of current ship design applied high efficient full spade rudder with bulb as energy saving. And, the result shows that these size vessels may have difficulty of meeting Phase 3 requirement because the service speed of these designs have already adopted current slow steaming concept and additional reduction of service speed is difficult in view of container shipping industry. In addition, the speed reduction may require more shipments, which could result in more CO2 emissions and the change of their shipping schedule. Thus, we need to consider carefully increasing reduction rate of this size of vessels compared to large size container vessels.

86 Annex 2, Page 32 Submitter Comments (with rationale) United Kingdom There are EEDI database indicating smaller containership (less than 49,000 DWT) could also meet the Required EEDI but is quite board but some still not. There would be worthwhile to look into containership at the lower end of less than 35,000 DWT such as size between 10,000 DWT and 20,000DWT which is likely remain in still not category. United States USG: The data from the EEDI database show that some small container ships do not meet the Phase 3 reduction rates at this time. However, the requirements apply only for new ships beginning in Because many small container ship designs already meet the requirements, it is reasonable to consider the reduction factors to be achievable overall. We think the data support retaining the 30% reduction factor for at least the smaller container ships; it may be reasonable to specify a greater reduction for larger container ships. It may even be possible to advance the effective date to 2022 or 2023, given the number of ship designs that already achieve the requirements. EC The new version of the EEDI database confirms the view the European Commission expressed in Round 1 that in general container ships can easily meet the phase 3 requirements. We note that concerns have been expressed regarding small container ships. The following detailed analysis however demonstrates that best-performers of container ships below DWT in phases 0 and 1 already meet the phase 3 requirements. In our view, phase 3 could and should be advanced to 2022.

87 Submitter Ship type/ size category Number phase 0 & 1 ships (mandatory EEDI) Comments (with rationale) Average performance of 10% best (in % below ref line) 'Early compliance' phase 3 Use of innovative MEPC 72/INF.12 Annex 2, Page 33 Container ship < DWT 57 41% 19% 0% CSC The EEDI scores in the latest release of the IMO EEDI database show that new containerships of 35-40,000 DWT built between have a mean EEDI 45% below the respective reference line (see Table 4); i.e. on average they are over-complying today by 15% to a standard they are not required to comply with for another 10 years time. The over-performance of the best 10% in this size category is even more impressive; at around 49% below the reference line. Table 4 Ship size categories Containerships in mandatory phases Average DWT of ships in size bin Mean EEDI Mean EEDI distance to reference line 30% 26% 41% 43% 49% 50% 45% Average distance to EEDI reference line of top 10% best ships 40% 46% 48% 56% 62% 64% 49% Source: Transport & Environment (Clean Shipping Coalition) CSC therefore, has difficulty understanding why the correspondence group is being asked to consider this issue as it is clear from the IMO s own EEDI database that containerships of this size are meeting and exceeding the requirements without difficulty. There are obviously some ships both in mandatory and non-mandatory phases that do not (yet) comply with the phase III requirement, but that doesn t mean it was impossible for them to comply. Some ships built between do not comply with Phase III because, at the very least, they are not required to by law, and economic drivers will not always lead to over compliance (e.g. split of incentives/landlord-tenant problem).

88 Annex 2, Page 34 Submitter Comments (with rationale) For this reason, it makes no sense to use the existence of some ships that are non-compliant with a future standard as an excuse for not tightening that standard. If all new ships in the fleet were to already meet the post-2025 (phase III) requirement, then there would arguable be no need for the EEDI requirement at all. The IMO process should be driving innovation and forcing the pace of efficiency improvements, not setting standards that are little more than business as usual market trends. The analysis of the top 10% best ships in each and every size class of containerships also indicate the potential to meet and overcomply, sometimes by a factor of 2, with the phase III requirements. This shows that containerships, both large and small can meet and indeed go beyond the phase III requirements. Therefore, CSC calls on the correspondence group should report this situation to MEPC72 in its interim report and recommend bringing forward Phase III compliance date from 2025 to This decision needs to be taken as soon as possible in order to provide clarity for ship-owners and shipyards alike. IACS ICS Ship size range between 10,000 DWT and 30,000 DWT is the real struggle since the lower threshold for required EEDI of containership is 10,000DWT, and in fact there is quite variability in attained EEDI up to 30,000DWT as shown on the graph of The EEDI database as of 31th of Oct ICS agrees that smaller containerships and feeder ships face a significantly greater challenge to achieve EEDI than large container ships. Whereas there is ample evidence to demonstrate that large container ships should be able to achieve EEDI phase 3, this is not the case for smaller containerships. This is largely because smaller feeder ships have tended to have lower design speeds than large ocean going ships. We would suggest that 40,000DWT could be considered as a threshold for additional measures to reflect challenges faced by smaller container ships. INTERFERRY INTERFERRY note that the smaller containerships may have difficulties meeting Phase 3, and just by the visual presentation of the available statistics for containerships, it is clear that <40,000DWT there is no apparent statistical correlation between DWT and attained EEDI. As suggested by INTERFERRY during Round 1 of this CG, if the underlying EEDI database shall be used as justification for any changes to the requirements, then proper statistical analysis must be made. Acceptance criteria should be developed for significance, deviation and size distribution. WSC An examination of smaller container ships ( TEU / 15,000 40,000 DWT) contracted after 2013 suggests that these ships should be able to meet the Phase 3 EEDI standards. Available data for the smallest container ships (999 TEU or less / < 15,000 DWT) contracted after 2013 was too limited at the time of our last review to draw conclusions. WSC will evaluate updated data on the container fleet in early 2018 to ensure that we are looking at the most relevant and up to date data applicable to the container fleet.

89 Annex 2, Page 35 Submitter Comments (with rationale) Whether a specific threshold is appropriate for the smaller container ships is a judgement that should follow a careful examination of the most relevant data available in the first quarter of 2018.

90 Annex 2, Page 36 Request 2-3 : Although a reduction of ship s design speed is not energy saving technology, this is one of measures to improve energy efficiency. Ships design speed are not always same for same ship type or ship size not only because of environmental reason but also economical reason. Thus, reducing ships design speed with maintaining the safety of ships could be considered and could have been taken as one of measure for improving energy efficiency of ships. Members are invited to provide the views on this point. If your answer for Round 1 is different when a reduction of ships design speed is considered as one of measures to improve energy efficiency, you could update your comment on a possibility of meeting Phase 3 criteria for each ship type and ship size. Further when a reduction of ships speed could be considered as one of measures for improving energy efficiency, how the difference of transport work could be taken into account through this review? Submitter Australia Denmark Comments (with rationale) No comment We do not see speed reduction as a real energy efficient improvement but only as a measure to fulfil/reduce the EEDI when you cannot do it with other technical measures. When we look at speed reductions especially for the slow ships, such as tankers and bulk carriers, we must be aware of fulfilment of the minimum power requirement, which has not yet been finally settled, of which reason we feel it too premature to discuss slow speed sailing for tankers and bulk carriers for the time being. We hope that we can agree on the final requirements such that these can be settled at the next MEPC meeting. Further If a speed reduction is introduced to achieve as certain EEDI value, this may remove the incentive for the designers and ship builders to improve the efficiency of the vessels. When discussing speed reductions we should not only look into the EEDI for the specific ship, but also on the energy efficiency of the complete transport work chain (or global transport work), having in mind that more ships are needed to keep a constant transport work per time unit. Germany Japan This means that the emissions per transport unit per time unit will not be a purely decreasing function of the speed. At a certain speed the auxiliary power, which is constant, will be so dominating that the energy demand per transport work per time rises due to the need for more ships to carry out the transport work, i.e. a clear minimum will appear at a certain speed below the normal service speed. Added resistance due to wind and waves will be relatively more dominating at lower speeds, which furthermore make the minimum energy demand issue more dominating. The choice of technical options for compliance with EEDI requirements should be kept as flexible as possible in order not to limit technical improvements in all fields. Hence, the design speed of ships should not be separately addressed in the EEDI regulatory framework. Design speed reduction, brought by using the engine with smaller MCR, should be considered as one of measures to improve energy efficiency of ships as long as these ships can maintain the safety performance. On the other hand, the effect of design speed reduction, such as increase in fleet volume and seafarers, should be taken into account.

91 Annex 2, Page 37 Submitter Netherlands Comments (with rationale) We see speed reduction as an operational choice, which at the same time can be a measure for improving energy efficiency of ships simply because the reduction of resistance at lower speeds in general is relatively larger than the reduction in speed. In general speed reduction could be an option to be considered by the ship owner for compliance with EEDI, however the ship owner could also choose for other options. It remains the choice of the ship owner/ operator. We are not sure if the question is to be understood such that if reduction of speed and therefore reduction of installed power to the level of what is required for the safe maneuverability in adverse conditions, could be an option to consider. If that is the case however, we have some observations that in our view need further consideration: 1. It may lead to rigid design rules which do not allow for flexibility; e.g. a ship owner may want to have some reserve dependent on the operational area of the ship; 2. It may lead to minimalistic designs as far as installed power is concerned; since adverse conditions primarily serves as a design condition and more severe conditions cannot be excluded this may lead to more accidents; 3. EEDI is not only about installed propulsion power and therefore this particular measure may lead to sub-optimisation since auxiliary machinery is not taken into account here; 4. It would also depend of the final guidelines for minimum power to be installed. As explained in 2-1 in our view these guidelines need further refinement. Moreover if such a measure would be considered we think expanding those guidelines to other ship types becomes more relevant. Norway As for the transport work issue when reducing speed to us it seems that if as an example speed is reduced by 10% and if the amount of cargo to be transported over a certain period of time remains the same, 10% more ships would be needed, thereby producing 10% more emissions (not taking into account possible effects of e.g. increasing ship sizes). Therefore we think the net effect of reducing ship speed is the reduction in installed power (in %) minus the reduction in ship speed (in %). This should be seen roughly speaking and in terms of thinking out loud. It neglects for example emissions in relation to the building of extra ships. It has been considered that reducing the speed is the best measure to improve ship efficiency. It is included in the current formulas for EEDI with the engine load at 75 MCR. When considering the reduction of ships design speeds, it is important to evaluate the overall environmental accounting. Some reflections that might question very low sailing speed for ships: By reducing the sailing speed of the smaller vessels, trucks could, for more trade routes, become a quicker way to transport cargo. Trucks are a less effective means of transport compared to ships when disregarding the very short distances. With a slower-sailing fleet more vessels might be required to meet the need for transport which s would affect the environmental accounting. Old vessels will be able to compete with new vessels as they are not restricted to EEDI.

92 Annex 2, Page 38 Submitter Republic of Korea Sweden United Kingdom Comments (with rationale) The initiatives of EEDI and EEOI are positive when it comes to emissions, but taxation of emissions would lead to even larger reductions. However, intercontinental sea transport has a long way to go in order to achieve a similar arrangement. No comments at this time. While an introduction of a statutory Design Speed reduction might potentially reduce the emitted GHG per shifted tonne mile, such a measure will neither stimulate more energy efficient ship designs nor the development of innovative energy saving. In other words, a speed reduction will not turn a bad performer into a good performer. Moreover, if transport needs are constant or even increasing then a reduced speed would eventually result in the need of additional potentially less efficient ships engaged in the transport work. When it comes to ro-ro cargo and ro-ro passenger ships it needs to be pointed out that these ships types are actually competing with land based road-hauled traffic. Hence, for these ship types, time table traffic and timely port calls are vital prerequisites in order not to lead to modal back-shift. Finally, safety issues related to potentially under-powered ships must be thoroughly assessed. Slow steaming running at the fraction of design speed/normal continue power (NCR) should not be considered as design technical improvement to improve energy saving. This practice is a function of operation and ambiguous for enforcement. United States USG: We should be very careful not to provide an incentive for shipbuilders to under power ships as a way to comply with the EEDI standard. EEDI is improved when the carbon-based fuel consumption is reduced. This can be achieved in a number of ways including improving energy efficiency of the engines and reducing power demand of the vessel on the engines. Power demand can be reduced through ship design (e.g. lower hull resistance, improved propellers, solar/wind power assist, etc) or through operational changes such as speed reduction. It has been well established that significant reductions in fuel consumption can be achieved through speed reduction during in-use operation. However, it is not clear how much speed reduction can be incorporated, in a meaningful way, into ship design. The EEDI equation recognizes power reduction, and the associated speed reduction as an improvement in EEDI. However, we need to be very careful on ensuring that this represents a real-world reduction in CO2, and not a paper reduction achieved through reducing the safety margin on power. The EEDI equation defines the reference speed as the speed that the ship travels when the propulsion engines are operated at 75% MCR in deep water under calm weather conditions. Simply reducing MCR of the engine will reduce the EEDI value correspondingly. However, if the practical result is that the ship operates at the same speed in-use as if it had larger engines, but at a higher fraction of engine MCR, then no real-world reduction is achieved. Additionally, the safety margin of available power is reduced. An idea has been discussed where the engine could have a safety mode, for adverse wind/tide/wave conditions, where the engine could change settings and operate at a higher power than rated MCR. If this is coupled with an engine governed for too little power for the ship, it could have the net effect of defeating the EEDI standards. In this case, the lower governor setting would seem to exist

93 Annex 2, Page 39 Submitter EC Comments (with rationale) solely for the purpose of lowering the MCR and speed at which the vessel is evaluated for attained EEDI. Therefore, an engine should not be allowed to have multiple power maps, chosen by the operator. Rather the full power range should be available to the operator at all times and a corresponding rated MCR be defined based on the full power range. We note we had some difficulties to fully understand the question; the use of ship design speed vs its effect on energy efficiency it s not clear to us. Is the question intention to explore the effects of slow steaming in the EEDI context (possible consequences to ship design?). Nonetheless, the following general comments are made in this respect: - In line with the Coordinator s summary of comments from round 1, reducing design speed should not be considered as technological development to improve energy efficiency, meaning its effect on design/technical efficiency be used in an operational context e.g. for transport work analysis. Instead, as some members highlighted, we should compare energy requirements (installed power) for the same amount of transport work to be done (cargo carrying capacity x distance sailed); installed power is the predominant factor/variable in the attained EEDI calculations. - There should be a clear distinction between design speed and service/operating speed or even the EEDI reference speed. Design speed should be understood as the value at which the ship will sail under its optimum hull/propeller/rudder efficiencies in the following conditions: at its design draught (usually summer load line), within a certain engine power margin and assuming the weather is calm (with no winds and waves). However, due to several factors such as fuel prices and freight rates, it is quite common that a ship that has been designed for 25 knots in practice does not move faster than 17.5 knots, the latest being considered as the service/operating speed but which already includes an additional power margin - the so-called sea margin, dependent from the region/s where the ship operates. But this is more from an operational performance perspective rather than a technical/design criteria and which would bring further complexity into de EEDI regulatory framework. Although both would contribute to energy efficiency improvements, one should split the technical/design efficiency from the operational performance. - It should also be noted that during the EEDI regulatory framework development, safety has always been considered a factor of paramount importance, leading not only to have included the minimum propulsion power requirements but also to had a decision to include the definition of V ref (reference speed) - which is taken at 75% of MCR, catering for the engine safety power margin (to avoid engine overload) and sea margin (to overcome adverse weather conditions, hull & propeller fouling, shallow waters, etc.). Furthermore, in the Guidelines on the method of calculation of the attained EEDI it is clearly stated that V ref, Capacity and Power should be consistent with each other, meaning that these variables should not be analysed in separate whenever the whole design efficiency is addressed; hull efficiency, propeller and rudder design are also to be taken into account. In summary, the European Commission believes that design/ technical efficiency measures (such as the energy saving within the EEDI) should be clearly differentiated - therefore not compared - from those related to the operational performance/ efficiency

94 Annex 2, Page 40 Submitter CSC Comments (with rationale) (such as speed reduction i.e. slow-steaming). Furthermore, further energy efficiency measures discussions are considered being out of scope of this EEDI review and out of the scope of the ToR CG. In its round 1 response, CSC provided information (MEPC 71/INF.13) showing little (less than 10%) to no reduction in the design speed (over the pre-eedi performance) of tankers and bulkers in different size categories (figures 1 and 2). This showed a non-linear relationship between engine power and design speed. Even though reductions in installed engine power can be used to attain a better EEDI, this does not necessarily mean a reduction in design speed because a better hull shape for example or the installation of 4th/5th term energy saving could reduce the need for higher engine power to deliver the same ship design speed, level of safe navigation and transport capacity under differing weather and load conditions.

95 Annex 2, Page 41 Submitter Comments (with rationale) Figure 1: Development of engine power, size and speed of Bulk carriers (2009=100% engine power, i.e. baseline) Source: CE Delft

96 Annex 2, Page 42 Submitter Comments (with rationale) Figure 2: Development in engine power, size and speed of Tankers (2009=100% engine power, i.e. baseline) Source: CE Delft

97 Annex 2, Page 43 Submitter Comments (with rationale) Furthermore, design speed is a characteristic of the ship, whereas transport capacity is a characteristic of the fleet. In few cases when changes in design speed could lead to a reduction in the fleet-wide transport capacity, markets would normally respond by increasing load factors, bringing idle/laid up ships into operation or building new ships. This would be the logical response of the market to rebalance itself. Therefore, CSC strongly recommends to keep capacity and transport supply questions out of the EEDI review process in order to avoid any misconception that higher design efficiency would create capacity bottlenecks in the global fleet. IACS ICS IACS would like to bring the issue of minimum power guidelines, which is yet to be resolved. In addition, operational speed reduction (operating ships below the design speed) is not efficient operation from the view point of hydrodynamics. (e.g. bulbous bow design) 1. The EEDI regulations are concerned with efficiency. Reducing speed will improve ship efficiency, notwithstanding our concerns at the implications for efficiency of very low speed operation, stated in paragraph above and paragraph below. The industry has already reduced speed, this is the primary reason for the excellent record of the industry in suppressing GHG emissions over the last decade despite increases in trade volumes. Different ship types operate at different speeds, those ships that operated at the highest speeds have achieved much greater reductions in their EEDI values than slower ships which is to be expected. Hence the disparity in the different rates of EEDI reduction seen between different ship types which risks the EEDI reduction rates of some ship types (e.g. bulk carriers) being misrepresented as failing to match the excellent performance of others (e.g. container ships). Speed reduction and reducing installed power is a valid means of improving efficiency provided that sufficient power is available to ensure that ships can maintain control in adverse weather conditions and in areas subject to strong currents. ICS considers that there is reason to believe this has been compromised in order to the improve EEDI value of some ship designs. As a result of these concerns ICS has commissioned an independent study which will evaluate the relationship between minimum power and maintaining sufficient ship control to assure safety for bulk carriers and tankers. ICS would again assert that we consider this to be a safety matter, not an environmental one. 2. On the question of further speed reduction and EEDI, further reductions in speed for ships such as bulk carriers would deliver much smaller improvements, if any, in efficiency than those already achieved by existing reductions in speed. In addition, there

98 Annex 2, Page 44 Submitter Comments (with rationale) is evidence to suggest that corrected transport work emissions (i.e. including adjustments for the greater number of vessels needed) will reach a minimum value, below which there is no benefit and that corrected transport work emissions may actually increase. This is as stated previously a result of wind and wave resistance become the dominating factor in resistance and energy use at very low speeds, not the propeller law relationship between power and speed. Auxiliary power is also relatively higher at very low speeds. 3. ICS considers that the level 1 method for assessing minimum power provided by the interim guidelines (MEPC.1/Circ.850/Rev.2) should be retained until further work can be undertaken to assess the degree of ship control which a ship should be able to maintain in adverse weather, the level of adverse weather within which a ship should retain this control and the minimum power necessary. This would entail placing on hold the work to complete guidelines for the level 2 assessment method until concerns over safety are fully addressed. As previously stated we do not believe that it is appropriate to agree to an early implementation of EEDI phase 3 until this work is completed as safety must always be the primary consideration of the IMO when developing and amending regulation. 4. ICS would highlight to the group our concerns regarding the risk that efforts to improve efficiency could result in ships being built with insufficient power for safety of navigation are not new and that these concerns have already been accepted by IMO as evidenced by the very fact that the 2013 interim guidelines for determining minimum propulsion power to maintain the manoeuvrability of ships in adverse conditions were developed. At MEPC 63 Greece proposed a minimum design speed as an alternative to defining minimum power (MEPC 63/4/15). This could have provided a simple and effective solution to address these concerns without necessitating the considerable research efforts made by SHOPERA and JASNOAE. Unfortunately this proposal was not accepted and it was decided instead to develop guidelines for minimum power. INTERFERRY Due to the vastly different design speeds within the ro-ro ship types (12-29 knots) and due to the paramount importance for high-value commodities to be able to be transported swiftly, the impact of design speed has to some degree been made less pronounced for roro cargo and ro-ro passenger ships, through the inclusion of fjroro. Ro-ro cargo and ro-ro passenger ships depend on short turn-around times and they often operate in busy shipping lanes where frequent speed variations are commonplace. There is therefore inherent need for redundant propulsion power, both to make up for temporary congestion in port and to provide maneuverability and flexibility during navigation. Therefore, using speed reduction as a tool to improve energy efficiency could not easily be done by reducing installed power. It would instead have to be approached in an operational manner, which is not compatible with the current EEDI framework. INTERTANKO We may not understand correctly the intent of the question. Since the CG deals with EEDI related matters, it seems there is a suggestion for a possible cap on the design speed or on the V ref for new ships. INTERTANKO will not recommend such an option. It will mean a further reduction on the installed power. To our knowledge, all EEDI certified tankers have all their main engines already de-rated. We used actual data from our members tankers and plotted the de-rated values of the 100% MCR installed power versus

99 Annex 2, Page 45 Submitter Comments (with rationale) the two minimum power lines given in the 2013 and 2015 IMO Guidelines, respectively. Most EEDI certified tankers have installed powers below the 2015 minimum power line and slightly above the 2013 minim power line. RINA As discussed in 2-1, Vref, Installed power and Pme are connected. The minimum power requirements govern installed power, and by default that means that Vref and Pme (75% of MCR) are also fixed if minimum power =installed power. RINA assumes that design speed=vref on summer load line, because otherwise design speed is not defined anywhere. Unless minimum power can be decoupled from Vref, either through the use of a different percentage of MCR, or by electronic means such as multiple engine maps, RINA does not see a possibility for reduction of Vref for tankers and bulk carriers.

100 Annex 2, Page 46 Submitter WSC Comments (with rationale) If we stay within the EEDI framework and ToR of this CG, mandating the reduction of design speed for other ship types means that existing ships would be able to operate faster than new ships, and there would be a push to try to operate new ship closer to 100% MCR to match the older ships; with a consequent increase in SFOC. This would mean that new ships would not be chartered. Any attempt to regulate speed would have to apply to both new and existing ships in order to avoid market distortion, however the methods and implications of such a regulation are more than can be addressed within a CG at this stage. It is clear that a reduction in installed power leads to a significant change in the EEDI value of a given ship. It is also clear that a reduction in installed power is one of the easiest mechanisms available to the yard and designer in meeting a given EEDI standard. Consistent with other comments made through the EEDI review process, greater attention needs to be devoted to the range of mechanisms available for improving the design efficiency of different ships. Simply relying on a reduction of installed power to achieve future EEDI standards is unsustainable and will likely lead to negative consequences in the efficiency and carbon footprint of the broader supply chain. WSC believes that a more fundamental and detailed examination of potential design modifications, their potential for improvement, and the commercial challenges that surround their uptake would benefit future discussions within the CG and the Committee.

101 Annex 2, Page 47 Request 2-4 : If members wish to make additional comment / to update their own comment on the possibility of meeting Phase3 criteria for each ship type and ship size, please provide it to this Group. Please bear in mind that all the comments have been shared to all members through this round as Annexes to this remark. It is not necessary to submit same kind of comment without having additional data/fact again. Submitter Comments (with rationale) Australia No comment Denmark No further comments at this stage Germany Japan Netherlands Norway No comments Based on the outcome of the case study, it is necessary for bulkers and tankers to reduce design speed by using the engine with smaller MCR or to apply additional energy saving, which are not usually used because of various limitation such as LNG fuel, for archiving 30 percent reduction of attained EEDI. This could be the same for other ship types. These changes would cause huge effect to the international shipping. Thus, these affect should be taken into account when considering the possibility of meeting 30 percent reduction of attained EEDI. In addition, it should be noted that unnecessarily stringent minimum power requirement will be a barrier to meet 30 percent reduction so as to achieve GHG emission reduction from new ships. We think the above answers make clear where we stand. Today, batteries, and thus hybridization, are not included in EEDI. Batteries can reduce energy use and emissions, e.g. through peak shaving or when manoeuvring. When calculating EEDI, only the energy carrier that has the highest burning value is included in the calculation. Batteries in a hybrid solution is therefore not included. This is a field that should be investigated further. Some information about hybridization: Batteries are still not treated in EEDI context The batteries can be charged and the stored energy can be used when the ship needs more power in e.g. adverse conditions Batteries can be charged either via PTOs on ME, WHRs on ME/AE, brakes for crane lowering, etc. Container vessels can achieve large efficiency improvements with peak shaving for refrigerated containers. Power can be used for e.g.: - Adverse conditions - Cargo handling - Extra power for higher speed - peak loads which normally need larger AEs to handle, or more AEs in operation - Fuel savings through running AE on/off and charging the batteries When PTO/PTI is installed, normally the PTO mode is used in EEDI calculations, because it is the most efficient mode and corresponds to the normal sea going condition, which EEDI is supposed to be used for. If you use a combined PTO/ PTI, the

102 Annex 2, Page 48 Submitter Republic of Korea United Kingdom EC Comments (with rationale) normal operating mode should be used in EEDI. This means you can install a large PTI for use in bad weather or to catch up delays without being penalized in EEDI. In today s formula, power take off (PTO) power take in (PTI) has a negative effect on the score. If using a PTI it can count it as a PTO if it is used as a PTO. A PTI/PTO solution can be used to reload speed. Boost at delay may be an additional effect when calculating the EEDI. No comments at this time. No additional comment. For oil tankers and containers ships, the number of phase 1 ships is sufficient to carry out statistical analyses comparing phase 0 and phase 1. The average reduction compared to reference line of phase 0 oil tankers is 26% and for phase 1 tankers 22%. For container ships, the average reduction of phase 0 ships is 40% and 38% for phase 1 ships. This indicates that the design efficiency of oil tankers and containers ships has been better in phase 0. The development of 'early compliance' (phase 0 and 1 ships already meeting phase 3 requirements) confirms this observation: 33% of phase 0 oil, tankers complied with phase 3 requirements, but only 15% of the phase 1 oil tanker. 79% of phase 0 container ships complied with phase 3 requirements, but only 67% of the phase 1 container ships This trend is a matter of strong concerns for us. The EEDI regulation should trigger efficiency improvements, not a deterioration of the design efficiency. This trend clearly demonstrates the need to raise the level of ambition of the present EEDI requirements. Advancing phase 3 to 2022 seems to be a logic consequence. We therefore suggest that the CG on EEDI review recommends to MEPC 72 to advance phase 3 to 2022, with the exception of bulk carriers above DWT and tankers above DWT. CSC A CSC commissioned study submitted to MEPC71 (ref) suggested that the design efficiency of new ships had plateaued in The EEDI scores in the latest release of the IMO EEDI database appear to support this conclusion in the case of containers and tankers (other ship types did not have enough phase 1 ships in the IMO database to perform the statistical analysis). Tables 5 and 6 below show that the average EEDI of container ships and tankers was worse in phase 1 than phase 0. The same worsening trend was also observed in relation to the share of ships already complying with Phase III requirements, as well as the best performing 10% of ships.

103 Annex 2, Page 49 Submitter Comments (with rationale) Table 5: efficiency of phase 0 ships Phase 0 ships ( , inclusive) Bulk Carriers Container Ships Tankers Total number of ships in 0 phase Average distance to EEDI reference line 20% 40% 26% Share with EEDI 30% under reference line 0.4% 70% 26% Average distance to EEDI reference line of top 10% 27% 58% 35% Source: Transport & Environment (Clean Shipping Coalition) Table 6: efficiency of phase 1 ships Phase 1 ships ( , inclusive) Bulk Carriers Container Ships Tankers Total number of ships in 1 phase Average distance to EEDI reference line 21% 38% 22% Share with EEDI 30% under reference line 0.0% 67% 14% Average distance to EEDI reference line of top 10% 27% 50% 33% Source: Transport & Environment (Clean Shipping Coalition) This suggests that unless the EEDI requirements are tightened, there is a risk that this backsliding could continue even to the point where design efficiencies fall back to those required by regulation. The significant gap between achieved efficiency levels and what is required by the regulation only underlines the urgency to ensure the requirements match and exceed the levels of efficiency that industry has clearly shown it is capable of achieving. Therefore, CSC strongly recommends to tighten phase 3 EEDI requirement for all major ship types by bringing the date forward from 2025 to 2022.

104 Annex 2, Page 50 Submitter IACS Comments (with rationale) (Coordinator s supplement: This comment was originally submitted as an answer to Question 1, but this was moved for readers convenience.) With regard to requested 1 (no specific answer sheet is provided), IACS has the following views: IACS is not in the position to provide this information, however would like to invite CG members to note that the following regulatory development may have adverse effects to EED: Use of scrubber (Global low Sulphur fuel from 2020) Tier III, NOx reduction compliance hardware in SCR or EGR Enhanced index R (Damage stability for passenger ships) - 1 Jan 2020 Impact out of black carbon (PM control) The following may affect attained EEDI depending on the direction of the discussion Electronic control of engine - mapping Engine test cycle - multi-purpose mass production engine (IACS UI MPC51(Rev.1)) ICS 1. Ships are trialing hull air lubrication systems (bubbling), increasingly effective energy recovery systems, hydrogen fuel cells, photo-voltaic cells, battery/electric hybrid power packages and with pure battery powered vessels being built for certain applications. LNG fuel becoming more popular. Flettner rotors are being installed on some ships and have potential to achieve a meaningful reduction in fuel use however they will not be suitable for all ships. The efficacy of all of these measures is determined as much by how they are integrated and ship operation as by any virtues of a particular technology. We have attached a summary of EEDI values for a series of modern bulk carriers and their EEDI values along with the operational problems experienced. These ships represent the most efficient designs which were available. 2. The opinion of representatives of the ship building sector which ICS has engaged with is that the only way to improve the EEDI value of slow, full hull form ships in the short medium term is switching fuel to LNG. This is because of the lower carbon factor of LNG and not because of an improvement in actual energy efficiency. Therefore the question of a fuel supply infrastructure is critical to any discussions of early implementation of EEDI phase 3 and potentially deeper EEDI reductions. Currently the supply infrastructure for LNG is very limited and is restricted to a handful of ports where LNG is particularly attractive to certain ship types, such as those which operate entirely within an ECA. These are not generally the ports used by bulk carriers and tankers, and there is little progress in developing an LNG supply chain in those ports

105 Annex 2, Page 51 Submitter Comments (with rationale) (or even regions) which serve the bulk tramp trades. This has significant implications for any proposals to advance the implementation of EEDI phase 3 or to amend the EEDI reduction rates and must be addressed before reaching any conclusions on EEDI early implementation or phase reduction. ICS would re-iterate that there should be no decision on early implementation of EEDI phase 3 without ascertaining whether the ship building sector believes this to be realistic and achievable for all ship categories subject to EEDI, and the market readiness of technological solutions which will be used to achieve EEDI phase 3 (eg. availability of alternative fuels such as LNG). 3. ICS is concerned that over optimising ship design to improve EEDI values risks degrading real world, in-service, efficiency of ships. This would negate the objective of EEDI, namely improving the efficiency of ships, an objective fully supported by ICS. INTERTANKO INTERTANKO would argue that even for smaller size tankers, the Phase 3 limit is not an easy target, since many of the current tankers, all with de-rated engines, modified hull lines and improved propeller design are still are below less than 30% of the base line. For 90k-120k DWT category, only 18% of Phase 0 tankers have an EEDI reduction below 30% from the base line, while out of 18 of the Phase 1 tankers on the same size category, only one has an EEDI value below 30% of the base line. A decision to move the application of Phase 3 ahead of 2025 may not be recommended until further in depth analysis is carried out. WSC Vessel characteristics and trends within a given fleet segment (i.e., ship type and size) will need to be considered as judgements are made regarding future standards. This may be necessary for a limited number of ship types and sizes with respect to the Phase 3 standards, but a more comprehensive assessment of technical and factual differences among different fleet segments may be useful when the Committee begins consideration of Phase 4 EEDI standards.

106 Annex 2, Page 52 Request 3-1 : There are some technical comments on the proposed correction factor for ice-classed ships. Members are invited to provide technical information and comment for these comments raised at Round 1. Submitter Australia Denmark Germany Finland Comments (with rationale) No comment No comments at this stage No comments. Japan and the Netherlands provided technical comments on the proposed new ice class correction factors for capacity contained in MEPC 71/5/6 submitted by Finland and Sweden. Our answers to these comments are given below. Comments of Japan: Table 3 of MEPC 71/5/6 contains average Cb values for less than 120,000 DWT, but there are ships of over-120,000 DWT cargo ships with ice class. It is necessary to include average Cb value for over-120,000 DWT ships. Our answer to the comments made by Japan: Our proposal was to determine the C b reference design separately for the five smallest ship subcategories given in table 3 (i.e up to and including Aframax size, see section 17 of MEPC 71/INF.16), because in our view Suezmax tankers and bulk carriers as well as very large tankers and bulk carriers are not very relevant subcategories for shipping in the Baltic Sea area, because they have limited access to the area due to the draft limitation in the Danish Straits. However, we agree that tankers and bulk carriers over DWT having an ice class do exist and therefore they should be included. Based on the analysis of ship data contained in the Maritime IHS ship data base for tankers and bulk carriers built since , in our view, the same values for C b reference design can be used also for these ship types for bigger ship categories than the Aframaz size. Therefore, our proposal is to extend the last ship category given in table 3 to cover also all ship categories bigger than the Aframax category, see the language in table 3 below highlighted in yellow colour. Table 3. Average block coefficients for bulk carriers and tankers for five size categories. C b reference design Ship type Small Handysize Handymax Panamax Aframax or a bigger ship category (< DWT) ( DWT DWT) ( DWT DWT) ( DWT DWT) Bulk carrier Tanker (> DWT)

107 Annex 2, Page 53 Submitter Comments (with rationale) Comments made by the Netherlands: In general we support the Finnish/Swedish proposal to replace the existing ice class capacity correction factors in the 2014 Guidelines with new factors and proposal to allow the use of the ship specific voluntary structural enhancement correction factor to calculate the ship specific ice class correction factor for capacity. In this way, the ice class capacity correction factor could be determined in a more accurate way for all ships having a Finnish Swedish Ice Class or any other ice class. We do however have 2 comments: 1. It seems so that the Fi factor for deadweight correction is not very accurate for small ships. Ref is made to the example ship of 3000 DWT as listed in the documentation TRAFI. Along the proposed formula for ice 1A: Fi = /DWT, the correction factor would be This means that the assumed deadweight loss (equal to the extra construction weight due to the ice belt steel) would be: (3000 * ) 3000 = tons According to the calculations made for each ship (TRAFI) the steel weight increase for ice 1A is 54 tons. Concluding: it seems so that for smaller ships the formula for Fi leads to overrating the deadweight penalty the small ice class ships have. 2. The methodology for capacity correction due hull lines adaption / displacement loss seems very rough if it is based solely on Cb. Cb is not only governed by bluntness of bowlines (affecting ice breaking). L/B ratio, B/T, transom immersion all have large influence on attained Cb but are hardly affecting icebreaking. Ships with high Cb (0.85) can be created with sharp entrance angles in the bow (favorable for icebreaking),in combination with large L/B. Comparing to a Cb of a reference design seems therefore very rough, and only fitted for a very slender band of ships with average main dimension ratio. Our answer to the comments made by the Netherlands (comment no 1 above): In document MEPC 71/INF.16 comparison of the new proposed ice class correction factors for capacity due to hull ice strengthening with the factors obtained by using the ship specific voluntary structural enhancement correction coefficient, f ivse, are made, see section 22 and table 8. In table 8 calculations are shown for the reference ships given in the appendix of the document having an ice class IA or IA Super concerning decrease of DWT due to ice strengthening. The results of the calculations are presented both for direct calculations (f ivse) and for the formulae given in table 1. Comparison of the results of the calculations for the direct calculation method with the results obtained by using the formulae given in table 1 indicate that the accuracy of the results obtained by the use of the formulae is sufficient for calculation of the ice class correction factor for capacity. We agree that the accuracy of the calculated weight

108 Annex 2, Page 54 Submitter Comments (with rationale) increase obtained by using the proposed new formulae compared to the calculated weight increase by using direct calculations varies to some extent, but in our view, the accuracy of the results obtained by the use of the proposed formulae is sufficient for calculation of the ice class correction factor for capacity. Finally, referring to our proposal given in the document MEPC 71/5/6, in the end of section 19, the capacity correction factor for ice strengthening of the ship (f i(ice class) ) can be calculated by using the formula given for the ship specific voluntary enhancement correction coefficient (f i VSE ) in section This formula can also be used for other ice classes than those given in table 2. This provision would allow the ice class correction factors for capacity due to hull ice strengthening to be calculated in a more accurate way. If there is a need to develop better formulae for the proposed ice class correction factors for capacity, we would need more data on the additional weight of the hull ice strengthening for small ships. We would welcome all such information! Our answer to the comments made by the Netherlands (comment no 2 above): In table 9 of the document MEPC 71/INF.16 a comparison of the existing and the proposed new ice class correction factors for capacity for ships having a high ice class is shown. More details of these ships can be found in Eronen and Riska (2014). Based on this information, in our view, the use of the parameter f icb, i.e. average block coefficient of a ship designed for sailing in open water only divided by the block coefficient of an ice-strengthened ship is a good parameter to measure the loss of capacity of the ice-strengthened ship. Figures 1 and 2 in MEPC 71/INF.16 also indicate that in ice-strengthened ships there is a significant loss of buoyancy in the bow area, if the bow is designed for ice-breaking purposes. It is of course true that other parameters like L/B ratio, B/T and transom immersion all have an influence on the attained C b, but according to the ice class rules the ship should have a certain ice-going capability, which may not be attained unless the bow of the ships is specially designed for ice-going purposes, which will generally lead to the loss of buoyancy in the bow area. Netherlands No further comments at this point in time. Japan No comment. Republic of No comments at this time. Korea Russian No comments now. Federation United No comment for this round. Kingdom United States The United States believes MEPC 71/5/6 offers a reasonable approach for further consideration and is an improvement over the present ice class correction factors. The issues noted by Netherlands and Japan in Round 1 will need to be considered before finalizing the correction factor fi. WSC No comment.

109 Annex 2, Page 55 Request 3-2 : If it is not necessary to rush into finalizing the correction factor fi, it is proposed to recommended to propose the Committee to consider fi and fj together at a later stage after having a proposal/proposals of correction factor fj. Do you have any views on this proposal? Submitter Australia Denmark Germany Finland Japan Netherlands Republic of Korea Russian Federation Sweden United Kingdom Comments (with rationale) No comment No comments at this stage No comments. We agree that it is very important to consider the combined effect of both correction factors f i and f j together at a later stage after having a proposal/proposals for the correction factor f j. We are planning to make a proposal for new ice class correction factors for power as soon as possible to this correspondence group. Both fi and fj can affect the design of ice class ships for meeting the EEDI requirement. Thus, Japan is of the view that consideration and amendment of fi and fj should be done at once. In our view the Fi approach for the extra construction weight is valid. In addition to the construction weight also the weight for the main engine and propeller shaft increase. We could support the approach to start the calculation with an average value and at a later stage you can use the Voluntary Structural Enhancement approach. The approach using the Cb does have some limitations and should be studied into more detail. We could support the proposal for considering fi and fj together at a later stage. We are in line with the coordinator s proposal. We support this approach. We agree that fi and fj should be considered together at a later stage. We can support to have a proposed fj (correction factor) for ice -classed ships first, then take the proposed fj to determine fi (capacity factor) at the later stage. United States The U.S. agrees that fi and fj should be considered together at a later stage after having a proposal/proposals of correction factor fj. CESA The Fi approach for the extra construction weight is a good one. Next to the construction weight also the weight for the main engine and prop.shaft rise. It is a good approach to start the calculation with an average value and at a later stage you can use the Voluntary Structural Enhancement approach. The approach using the Cb does have some limitations and should be studied into more detail. The latter would mean that it is a good proposal to consider fi and fj together at a later stage. WSC No comment.

110 Annex 2, Page 56 Request 4 : Members are invited to provide detailed data and analysis for the possibilities of applying the energy saving listed in Annex 2 and Annex 4 to this remark for ships having ice class. The data and analysis for each ice class would be useful for further consideration. Submitter Australia Denmark Finland Comments (with rationale) No comment No comments at this stage We have the following general comments on the possibilities of applying certain energy saving listed in Annex 2 and Annex 4 for ships having an ice class: 2 Recovery of propeller energy (1) Coaxial contra-rotating propeller: Coaxial contra-rotating propellers have not yet been applied on ships having an ice class due to problems related to ice strengthening. Higher ice-propeller interaction is also expected to occur due to two propellers. (2) Free rotating vane wheel: Free rotating vane wheels have not yet been applied on ships having an ice class. Probably not suitable for ships having an ice class, if this device is expected to be damaged even under rough sea conditions in open sea. (3) Ducted propeller: Can be used for ships having an ice class even if there is a risk of clogging the nozzle by ice in certain ice conditions. (4) Pre-swirl devices: Issues related to ice strengthening may restrict the use of this device in ships having an ice class. The Mewis duct might be clogged by ice in ice conditions. (5) Post-swirl devices: Issues related to ice strengthening may restrict the use of this device in ships having an ice class. Postswirl devices are counterproductive when the ship is reversing in ice. (6) Split stern: This is in use on ships having a high ice class. 7 Other (4) Hull vane: Not suitable to be used in ships having an ice class. Germany No comments. Japan No comment Netherlands No further comments at this point in time. Republic of No comments at this time. Korea Russian Analysis for the possibilities of applying the energy saving see table in annex. You can see from the table that most of Federation the proposed cannot be effectively applied to the ice class ship without harm to safety or cannot be applied at all. United No comment for this round. Kingdom United States No comments at this time.

111 Annex 2, Page 57 Submitter Comments (with rationale) INTERTANKO The energy saving which have already normally used are not the which could further improve the energy efficiency of ships. WSC No comment.

112 Annex 2, Page 58 Request 5-1 : Following is one of possible proposal of defining ice class ships higher than IA super. Members are invited to provide a comment on the following draft exemption of EEDI regulation. Ice class ships having at least an ice class PC 5 of IACS or equivalent (see regulation and of the Polar Code) are exempted from EEDI regulation. Submitter Comments (with rationale) Australia Australia supports the definition above for ice class ships higher than IA Super as an addition to the existing regulation 19 exemption in Annex VI to MARPOL. Finland We can agree to this proposal. Germany No comments. Japan No comment Netherlands No further comments at this point in time. Republic of No comments at this time. Korea Russian Russian Federation supports the definition above for ice class ships higher than IA Super as an addition to the existing regulation Federation 19 exemption in Annex VI to MARPOL. Sweden We can support the suggested exemption. United States The United States notes that in the Polar Code, a category A ship is defined as a ship designed for operation in polar waters in at least medium first-year ice, which may include old ice inclusions. This category of ship should be exempted from EEDI regulations. Category A ships would include ice class PC 5 and above of IACS Unified Requirements or equivalent. Additionally, any recommendation for exemption should make reference to IACS PC 5 and above or equivalent, in lieu of explicit reference to any individual rules. This approach would be consistent with the approach used in the Polar Code. Therefore, the United States can agree with the suggested proposal. WSC No comments.

113 Annex 2, Page 59 Request 5-2 : Members are requested to provide comments on the table contained in Annex 6-2 which is provided to this group for consideration. Submitter. Australia Finland Comments (with rationale) Australia considers that it would be useful for all classification societies to adopt the International Association of Classification Societies Polar Class notation where applicable. We have provided two minor editorial improvements to the table contained in Annex 6-2, one for the Finnish-Swedish Ice Class Rules and one for the ice class rules of ABS. We have also the following general comments on this matter: - The information contained in Annex 6-2 is in line with our own equivalence tables for ice classes equivalent to the Finnish- Swedish ice classes (see -> Maritime -> Ice classes of ships -> List of equivalences, 2017), which are based on our analysis. - The information contained in Annex 6-2 is also in line with the information contained in the annex to the HELCOM Recommendation 25/7 (see -> Recommendations -> Maritime). Russian Federation IACS Finally, if ice class ships having at least an ice class PC 5 of IACS or equivalent (see regulation and of the Polar Code) will be exempted from EEDI regulations (see request 5-1), it might be relevant to develop such an equivalence table for ice class PC 5, where ice classes equivalent or higher to PC 5 would be listed. Russian Federation considers that it would be useful for information only. We agree with IACS comments that a complete comparison of ice classes is impossible due to the existence of different ways of solving identical tasks related to the operation of ships in ice and complete equivalence for all requirements regulated by different ice classes is not possible. But for EEDI purposes the comparison can be simplified to comparing the ice conditions in which the operation of the ice ship is allowed. The purpose of the proposed table is not a comparison of classes, but a list of ice classes to which the requirements for EEDI should not be applied because the additional power of the main engine is necessary to overcome the ice conditions in which the operation of the ice class ship is permitted. With regard to request 5-1 and 5-2, IACS has the following generic views. In the previous round, we had provided the following comments: IACS was unable to discuss this issue in depth owing to the late participation in the CG. The following view was expressed by one member (not yet fully agreed by IACS members) but may provide basis for discussion in the future round of CG. IACS will provides its consolidated comment at a future round.

114 Annex 2, Page 60 Submitter. Comments (with rationale) Ice class IA-Super ships have such structure, engine output and other properties that they are normally capable of navigating in difficult ice conditions without the assistance of icebreakers. The design requirement for ice class IA- Super is a minimum speed of 5 knots in the brash ice channels of (thickness at mid channel) HM = 1.0 m and a 0.1 m thick consolidated layer of ice. Requirements for ice class higher than IA-Super could include e.g. a speed of above 5.0 knots in ice channels with HM > 1.0 m and consolidated layer of ice with thickness exceeding 0.1 m. After some discussions in this round, we would like to offer the following comments as a matter of principle: It should be pointed out that we are not defining ice class ships higher than IA Super, but defining, as I understand it, ships of a higher ice class the power requirement of which means that achieving EEDI becomes impossible. Any ice class is not equivalent and the table is not definitive equivalency in respect of different Class requirements to ship s hull structure and power. The table can be used as reference to which ice class notations are high than 1A Super at present. Acceptance in the Baltic (HELCOM) such as material requirements etc. is different. Rather than referring to ice class (to which IMO does not have control), referring to polar code (category A ships) may be an idea. ***

115 Annex 3, Page 1 ANNEX 3 SUMMARY OF LATEST INFORMATION OF ENERGY SAVING TECHNOLOGIES *1 Short-description of each can be found in the original table. *2 The information on cost of each is just for reference and is not directly related to the availability of. ratio when The year Categorization using with expected for inuse of that of application application Current status Written comments related to the possibility of / Type of other energy ratio (%) saving technology (%) 1 Design improvements (1) Optimization of hull 0-3% The improvement potential of of category 3, 6(3) and 7(1) might be slightly reduced. Expected improvements from holistic design optimisations (hydrodynamic hull form only) are expected to achieve full 5% - 7% improvements for the entire operational envelop. Already available Normally used This technology has been improved for a long time and there are few room remained for further improvement. Please note that optimization of the hull is reflected in the Vref (speed) at scantling draft on the chosen power. This draft is however for many ship types rarely utilized. If optimizing the hull lines for this draft to achieve a better EEDI, the performance on actual drafts may not necessarily be optimum. By focusing much on hull design for new buildings compared to standard yard design you can probably gain the indicated. The effect of the optimization of the hull can be limited by the possible following reduction of the cargo capacity. This can result in bader performance measures in CO2 per transport unit. Stern wedge is only relevant for fast vessels. In today slow steaming it is not relevant. Estimated whole cost (US dollar) *2 0 to hundreds of thousands of dollars

116 MEPC 71/INF.12 Annex 3, Page 2 Categorization / Type of (2) Optimization of superstructure (3) Optimization of propeller (4) Podded (azipod) drives ratio (%) 0-3% (5-30% reduction of wind resistance.) ratio when using with other energy saving (%) 0-3% This could interfere with other energy saving devices especially for devices fitted on stern part of ship. The improvement potential of category 7(1) might be slightly reduced. Combination with category 3 is not applicable. Similar to the effect of conventional propeller. The improvement potential of category 7(1) The year expected for inuse of that technology Available by 2020 (partly already achieved.) Already available Current status of application Sometimes used with specific ship types, e.g. car carrier, high speed crafts. Normally used Rarely used Written comments related to the possibility of application It is questionable if it will make a difference in streamlined design of super-structure for a container vessel as it will often be completely covered by container stacks. A spoiler design on bow (cover) might be helpful on smaller ships. Operational you gain by better distribution of containers. However for tankers and bulkers it will be possible as we also se it for car carriers On 2-stroke engine is it not relevant with reduction gear. We have not seen high savings for new propeller types like CLT or Kappel. In a few cases, we have seen 1-2% saving compared to a traditional propeller design with latest technology. However, both CLT and Kappel often suffer from Cavitation. Many discussions on scaling, but even full scale test has not showed premium performance. Increase of diameter (low rpm) are a more robust solution. This will lower Cth and reduce benefit for propellers designed for high loading. Large propeller diameters can result in problems with cavitation. Competing goals, such as noise reduction or erosion-free long-term operation, might reduce the reduction potential This could be effective only for ships which change the course frequently such as tugs, supply vessels and other vessels for short sea shipping. It is not effective for other ship types. Estimated whole cost (US dollar) *2 0 to hundreds of thousands of dollars 0 to hundreds of thousands of dollars (The cost depends on the size of engine and propeller.)

117 MEPC 71/INF.12 Annex 3, Page 3 Categorization / Type of (5) Matching and integration of hull form and drivetrain ratio (%) 2 Recovery of propeller energy (1) Coaxial contra-rotating propeller ratio when using with other energy saving (%) might be slightly The year expected for inuse of that technology reduced. 1-20% Already available 8% This technology collects the energy of rotational flow. This is why this could interfere with other energy saving devices which have same effect such as pre-swirl devices and post-swirl devices. The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Current status of application Sometimes used Application is limited due to its expensive cost. Written comments related to the possibility of application Competing goals, such as noise reduction, might reduce the reduction potential Mostly relevant for highly loaded propellers, e.g. for vessels with draught restrictions. Complicated and expensive installation and maintenance. Expected savings to be in range 0-5% Future application would be limited because of its application cost. Estimated whole cost (US dollar) *2 Millions of dollars

118 MEPC 71/INF.12 Annex 3, Page 4 Categorization / Type of (2) Free rotating vane wheel ratio (%) ratio when using with other energy saving (%) Combination with category 2(2) is not applicable. 0-10% Will not work together with other post-swirl devices like PBCF, Costa Bulb, Asymmetric rudder The year expected for inuse of that technology Already available Current status of application Rarely used Written comments related to the possibility of application 10 percent would be too large. Therefore, 10 percent should be considered as maximum improvement ratio. No experience. Will saving be related to loading of propeller. With large diameter propeller, this is not found to become relevant Estimated whole cost (US dollar) *2 To fit large diameter propeller could cause negative affect to the improvement of hull form. Competing goals, such as noise reduction, might reduce the reduction potential This device is not applied recently because there could be a damage to this device under rough sea. The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category

119 MEPC 71/INF.12 Annex 3, Page 5 Categorization / Type of ratio (%) ratio when using with other energy saving (%) 2(2) is not applicable. The year expected for inuse of that technology Current status of application Written comments related to the possibility of application Estimated whole cost (US dollar) *2 (3) Ducted propeller (4) Pre-swirl devices 0-10% The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. 1-4% ( 4-6% improvement would be achievable in case post-swirl device in 2.(5)) is used together) This could interfere with CRP or other pre-swirl devices. The effect of this technology depends on stern hull form. This device collects stern Already available Already available Rarely - sometimes used Sometimes - normally used The improvement ratio depends on the thrust loading condition. Only useful for ship types with high propeller loading. 10 % seems a little high however good experiences on tug boats Not relevant for container vessels This device could be effective for ships with high propeller load. This has not been applied for conventional cargo ships. The improvement ratio is reduced because of interfere of devices. Full scale tests on an optimised PSS indicate 7% improvement (PD) for a bulk carrier (ref. EU GRIP project, ) Our experience with PSS (DSME) and with BTF (Becker) are predicted saving in range 2-4%. We do not have any full scale PSS installation, but other owners have experienced structural problems. Tens of thousands to hundreds of thousands of dollars

120 MEPC 71/INF.12 Annex 3, Page 6 Categorization / Type of ratio (%) ratio when using with other energy saving (%) viscous vortex, and this is why the effect is larger for fat ships than slender ships. The year expected for inuse of that technology Current status of application Written comments related to the possibility of application We have no full scale performance records for BTF. Previous experience with Mewis duct on VLSS showed a saving of 4-5% which was confirmed in full scale. Mewis ducts are not suitable for Container vessels. This is not used for high-speed ships. Estimated whole cost (US dollar) *2 (5) Post-swirl devices 1-4% ( 4-6% improvement would be achievable in case pre-swirl device in 2(4) is used together) The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable This could interfere with CRP or pre-swirl devices. The effect of this technology depends on stern hull form. Already available Sometimes - normally used Savings of 2-4% for a well-designed asymmetrical with rudder bulb have been seen. Requires a propeller cap that closes the gap to rudder bulb. Savings from PBCF between 0-2% are found. Tens of thousands to hundreds of thousands of dollars The improvement potential of

121 Categorization / Type of (6) Split stern (Twin skeg) ratio (%) ratio when using with other energy saving (%) combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. 0-6% The improvement potential of combinations within category 2 and with 1(1), 7(1) might be slightly reduced. 3 Engine energy recovery (1) for lowspeed engine with category 2-6% Combination 4(2)(4) is not applicable. The year expected for inuse of that technology Already available Already available Current status of application Rarely - sometimes used Sometimes - normally used Written comments related to the possibility of application Potential in application highly depends on the application case. Biggest gains can be achieved if right combination of engine can be installed in order to have a large propeller diameter as possible. 4 6 % realistic however require a big (and expensive) CAPEX Effective for ships with low depth and high power This could be applied only for ultra-wide ships. This is not suitable for conventional ships. Application comes with the risk of poor wakefield and related problems with cavitation, vibrations, shaft bearings. Efficient combination with EGCS The effect of this device could be taken into account as reduction of PAE. Maximum effect of this would be about 5%. The effect of this technology would be smaller if temperature of exhaust gas is relatively low. MEPC 71/INF.12 Annex 3, Page 7 Estimated whole cost (US dollar) *2 Cost would be significantly increased because of complex stern hull form. Hundreds of thousands to millions of dollars

122 MEPC 71/INF.12 Annex 3, Page 8 Categorization / Type of (2) for mediumspeed engines (3) for gasfuelled engines ratio (%) 3.5% of shaft power. 5% by turbo Generator; and 10% by turbo Generator + Power Turbine 3.5 % seems high. Reduction of CO2 22% (including 13% energy shaft power recovery) ratio when using with other energy saving (%) Combination with category 4(1) is not applicable.. Combination with category 4(1)(2) is not applicable.. The year expected for inuse of that technology Already available Already available Current status of application Written comments related to the possibility of application Estimated whole cost (US dollar) *2 This device is applied for larger ships which are suitable for this device. Sometimes used Efficient combination with EGCS Millions of dollar The effect of this would be smaller if temperature of exhaust gas is relatively low. This technology is applied for specific ship type. To fit this would be difficult because of its arrangement. The effect of this device could be taken into account as reduction of PAE. Maximum effect of this would be about 5%. Rarely - sometimes used Infrastructure for LNG fuel should also be provided. There are few infrastructure available in the world at this time. More than millions of dollars (4) Machinery arrangement and hybrid propulsion 5% by turbo Generator; and 10% by turbo Generator + Power Turbine CRP-Pod vessel could improve the propulsion performance by 13% compared with the conventional twin shafts and The effect of this would interfere with pre-swirl and post-swirl device Combination with category Already achieved Rarely used - normally used with suitable ship types In general, CRP is more effective than CRP-Pod. Thus this technology is only applied for high-speed ROPAX and other limited ship type. The effect of this technology would be lower.

123 MEPC 71/INF.12 Annex 3, Page 9 Categorization / Type of ratio (%) propellers system. ratio when using with other energy saving (%) 4(1) is not applicable. The year expected for inuse of that technology Current status of application Written comments related to the possibility of application Estimated whole cost (US dollar) *2 (5) Fuel-cell propulsion (6) Boiler efficiency with Waste Heat Recovery (Economizer) (7) Optimization of Exhaust gas via Waste Heat Recovery ( Main & Aux engine) (5-10%)* 25-30% (100%) * aux. engines only in short sea shipping with hydrogen produced with renewable energy 30% (VFD) more efficient at a firing rate of 80% WHR plus VFD = 5% increase in efficiency (30% +5%) 3.9 to 5.6% Up to 15% when used in conjunction with a condenser Under development 2020* Already available VFD not chosen because of higher cost. Already available Never used for conventional commercial vessel trial application on-board ferries, container ships and offshore vessels, Cruise ships in 2018/2020 Sometimes used Normally Used for main engine waste heat recovery Sometimes used for auxiliary engine Today only systems with small power output available There would be difficulty of application because of cost and size. Timely finalization of IMO regulation and availability of appropriate FC primary fuels, such as methanol, low-flashpoint diesel or hydrogen Cost for VFD is between US$ to US$ excluding installation. Waste heat recovery standalone 25,000.00US plus 10,000US for the electric generator

124 MEPC 71/INF.12 Annex 3, Page 10 Categorization / Type of (8) Exhaust aftertreatment solutions (SCR) ratio (%) ratio when using with other energy saving (%) 3-5% This can be combined with other. The year expected for inuse of that technology Already available Current status of application Written comments related to the possibility of application Already increasingly used for Tier III NOx compliance; Rarely used for EEDI improvement Application as Tier III NOx compliance only for new ships slow market roll out Estimated whole cost (US dollar) *2 CAPEX 0% when used with an engine for Tier III application 3-10% of engine costs when specifically mounted for EEDI improvement OPEX Impact to OPEX depend on correlation of fuel price and Urea price Due to improved SFC, a range between cost neutrality and up to 3% OPEX reduction is possible

125 MEPC 71/INF.12 Annex 3, Page 11 Categorization / Type of 5 Hull coatings (1)Selection of coatings (2)Polymers and air lubrication ratio (%) 1-2% reductions of frictional resistance by Low friction coating The effect of this could be taken into account as the difference of Vref at sea trial. This is why the effect to attained EEDI is only 1-2 %. ratio when using with other energy saving (%) The effect would be reduced when using other friction reducing devices The improvement potential of combination with 5(2) might be slightly reduced. 1-3% The effect would be reduced when using other friction reducing devices The improvement potential of combination with 5(1) and 7(1) might be slightly reduced. The year expected for inuse of that technology Already achieved. Already achieved. Current status of application Sometimes - normally used Hardly a design feature. But sure right coating will reduce fuel. Rarely used Written comments related to the possibility of application Tightening of environmental policies regarding permissible biocides resulting in lack of efficient coatings The improvement ratio is different according to the ship type; higher improvement ratio could be expected for a blunt ship such as a tanker and a bulk carrier, than a slender ship such as a container carrier. Previous experience with micro bubbles did show a very small net gain. We are looking into updated. Predicted propulsion savings of some 4-5%, but a loss for air bubbles production will eat from savings. Surface treatments need to be mechanical robust and have good anti fouling properties such any gain are sustainable. Think expected savings are on high side. Estimated whole cost (US dollar) *2 Tens of thousands to 1.5 million dollars. (There are various type of coating with different cost and effect.) Millions of dollars

126 MEPC 71/INF.12 Annex 3, Page 12 Categorization / Type of ratio (%) 6 Alternative fuels and energy (1)Liquefied natural gas (2)Biofuels ratio when using with other energy saving (%) The year expected for inuse of that technology 20% - 25% Currently in-use where LNG is available Already available (3) LPG % Already available (4) Alcohol Methanol Recently 7-10% developed (5)Wind power/ Auxiliary wind propulsion Ethanol 3-5% Flettner Rotor type applications in good wind This can be combined with some of the Already available Current status of application Rarely - sometimes used Rarely used Rarely used Rarely - sometimes used Sometimes used on cruise ships, Written comments related to the possibility of application Based on the current technology, this is only effective to ships with wider and shallow hull. The effect of air lubrication would be reduced if flat part of bottom is narrow or depth is large. Deeper depth requires more energy for providing air to the bottom. Concerns on methane slip and infrastructure World-wide implementation is governed by the growth of supply infrastructure and harmonization of shore-side safety requirements. Infrastructure for LNG fuel should also be provided. There are few infrastructure available in the world at this time. Availability of fuel These fuels might require adaption of engine technology and/or fuel treatment, handling and storage procedures on board Marine regulations for safety requirements to be adapted and to be finalized promising option for LPG tankers; amendment of the IGF Code required Infrastructure for methanol fuel should also be provided. There are few infrastructure available in the world at this time. Promising option for all shiptypes because fuel is liquid and can be stowed onboard in integral tanks. Amendment of the IGF Code required Not relevant for container vessels due to lack of deck space. Estimated whole cost (US dollar) *2 Millions to tens of millions of dollars For rotor applications Flettner

127 MEPC 71/INF.12 Annex 3, Page 13 Categorization / Type of (6)Solar power ratio (%) conditions can achieve approximately 5%/unit long term average fuel savings in typical bulker and tanker operations. 5-15% for ship types with available deck space Based on case studies for a big tanker and bulker, Flettner Rotors reduce EEDI by 6-7%. 1% reduction of power consumption for ship types with sufficient deck space available ratio when using with other energy saving (%) other without loss of improvement ratio, in particular categories 5, 6, and 7. The improvements potential of of category 1 and 7(1) might be slightly reduced. The engine / fuel related improvement potential remains, but applies to reduced installed power. potential will increase in combination with (integration into) sails due to the increase of area The year expected for inuse of that technology Already available Current status of application ferries, ro-ro ships and yachts Rarely used Written comments related to the possibility of application no obstacles for auxiliary use of wind power; sail only propulsion difficult due to safety concerns Not relevant for container vessels as only very little deck area are available. Applicable ships are limited because of their arrangement. This technology has been used just as trial basis. Because of the limitation of space, maintenance and cost effectiveness, this would not be applied for conventional ships. Estimated whole cost (US dollar) *2 400k-1M/unit when installed onboard, depending on model Millions of dollars

128 MEPC 71/INF.12 Annex 3, Page 14 Categorization / Type of ratio (%) 7 Other (1)Optimization It depends on of principle development of dimensions infrastructure. (2)Reduction of light weight (3)Hull Vane ratio when using with other energy saving (%) available for panels The improvement potential of of category 1 and 7(1) might be slightly reduced. The engine / fuel related improvement potential remains, but applies to reduced installed power. The year expected for inuse of that technology Already available 0-3% Already available The improvement ratio depends on the ship types and the speed. It is reported that on merchant Slightly reduced potentials can be anticipated in combination with category 2. Current status of application Normally used Written comments related to the possibility of application Most likely to many constrains. However, we will often look at main dimensions to find lowest transport cost. It depends on development of infrastructure. Estimated whole cost (US dollar) *2 None Normally used Wider use of innovative lightweight materials, such as None FRP, require revision and modification of IMO instruments The effect of H-CSR should also be considered. The effect of this technology depends on ship type and ship size. Already in use. Rarely used Hardly relevant for bigger vessels due to slow steaming.

129 MEPC 71/INF.12 Annex 3, Page 15 Categorization / Type of ratio (%) ships, the potential resistance reductions between 5% and 10% are common. ratio when using with other energy saving (%) The year expected for inuse of that technology Current status of application Written comments related to the possibility of application Estimated whole cost (US dollar) *2 ***

130

131 Annex 4, Page 1 ANNEX4 LATEST INFORMATION OF ENERGY SAVING TECHNOLOGIES *1 The information on cost of each is just for reference and is not directly related to the availability of. ratio when The year Categorization using with expected for / Type of Short-description other energy ratio (%) in-use of that saving technology (%) Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) 1 Design improvements In order to achieve reduction of drag (resistance) through optimization, it is necessary to find adequate approaches which will ensure the validity of optimal design from a global perspective, allowing detailed and refined optimization of hydrodynamic design. On the other hand, an alternative approach which allows for complete optimal design is also possible. The main difference of this kind of approach in respect to the first one lays in the ability to define several conflicting tasks, and yet arrive at an optimum solution which best suits optimal ship design in respect to her involvement in a specified operational mode. In spite of the fact that the second approach clearly shows advantages, it should be emphasized that the methods that are applied mostly rely on CFD modelling, and consequently on the experimental results from a towing tank that are used, amongst other purposes, also for validation of theoretical results from CFD. In general, hydrodynamic optimisation must address the full range of operational conditions (multi-point), hence a complete or holistic design optimisation approach including also wind and waves (added resistance) should be favoured. The percentage of new designs that are subjected to systematic optimization of the hull and of the propeller compared to the percentage of designs that are built merely on the basis of existing experience is currently unknown. However, in general, it is believed that probably the greater proportion of new designs today are going through some systematic form of optimization of hull and propeller design, focusing on reduced resistance (drag reduction) and increased propulsive efficiency. (1) Optimization of hull There are many barriers to focusing solely on modifying the hull lines to achieve more favourable resistance. Examples are the effects of the given requirements of the amount and type of payload and the dimensions of ports and terminals. These barriers will considerably reduce the potential for the reduction of resistance and of fuel consumption. On single ships, improvements in power requirements of up to 30% have, in fact, occasionally been achieved on particularly ill-conceived designs; however, the mean potential for improvement would be expected to be small. However, some potential exists in the optimization onto operational profiles instead of a single design condition. Smaller ships are more sensitive to design details, since they have comparatively large wavegenerating resistance and also because less resources will traditionally be available for optimization, due to the 0-3% This technology has been improved for a long time and there are few room remained for further improvement. Please note that optimization of the hull is reflected in the Vref (speed) at scantling draft This can be combined with most of the other without loss of improvement ratio, in particular categories 1, 2, 4, 5, 6(1)(2) and (4) and 7(2). The improvement Already available Normally used By focusing much on hull design for new buildings compared to standard yard design you can probably gain the indicated. The effect of the optimization of the hull can be limited by the possible following reduction of the 0 to hundreds of thousands of dollars

132 MEPC 71/INF.12 Annex 4, Page 2 Categorization / Type of Short-description smaller overall budget that is typically available for developing the designs. The expected improve ratio depends on an individual ship. Resistance and energy consumption increase during ship operation in a natural seaway. Traditionally, ships have been optimized primarily for calm-water conditions in a towing tank (not least because the contractual measurements of trial performance are conducted in calm water); however, optimization for irregular wave conditions is becoming more common. During their lifetime, ships will more frequently operate in a wave field that is characterized by the short wavelength λ (small sea states) in comparison to the ship length L. Therefore, optimization for waves generally emphasizes shortwavelength waves (this however holds only for large vessels, smaller ships (feeders, coastal ) do operate in conditions λ/l 1). There is an example in relation to the technology for the optimization of hull, namely, Stern Wedge. Example 1 Stern wedge By guiding the propeller flow a bit down a lift is created in the aft part of the hull which can reduce the engine power. The optimum slope of the wedge is often appr. 10 degrees. This type of improvement is normally only beneficial for slender and relatively fast ships (typical navy ships and fast patrol vessel or similar) A power reduction of up to 10 % can be obtained. ratio (%) on the chosen power. This draft is however for may ship types rarely utilized. If optimizing the hull lines for this draft to achieve a better EEDI, the performance on actual drafts may not necessarily be optimum. ratio when using with other energy saving (%) potential of of category 3, 6(3) and 7(1) might be slightly reduced. Expected improvements from holistic design optimisations (hydrodynamic hull form only) are expected to achieve full 5% - 7% improvements for the entire operational envelop. The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology cargo capacity. This can result in bader performance measures in CO2 per transport unit. Stern wedge is only relevant for fast vessels. In today slow steaming it is not relevant. Estimated whole cost *1 (US dollar) There are some applicable for certain types of ships.

133 MEPC 71/INF.12 Annex 4, Page 3 Categorization / Type of Short-description Superyachts are generally characterized by Loa between 40 and 100 meter, high Froude numbers (>0.35) and slender lines (CB<0.5). Most yachts are full displacement vessels; a limited number is (semi) planning. Motion behaviour is an important design criterion in connection with required levels of comfort. Therefore hullform innovations are generally not only aiming at reducing resistance but also improving sea keeping properties. However, recent hullform innovations result also in (considerable) reductions of hull resistance, not only in adverse sea conditions, but also in still-water conditions. Recent hullform innovations for superyachts comprise: Axe bow ( and J.L. Gelling, "The Axe Bow", HISWA Symposium 2006) Supersport hullform ( ) Fast Displacement Hull Form ( and Perry van Oossanen et al, "Motor Yacht Hull Form Design for the Displacement to Semi- Displacement Speed Range", FAST 2009 Conference) All above hull forms are applied on newbuilds of SYBAss members and are currently in service. For confidentiality reasons we have provided publicly available information only (refer to the above websites and reference papers). The are also applicable for other ship types of comparable characteristics and/or type of operation, e.g.: the axe bow has also been applied on a number of ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar)

134 MEPC 71/INF.12 Annex 4, Page 4 Categorization / Type of (2) Optimization of superstructure Short-description fast crew suppliers, windfarm maintenance vessels, fast patrol and Search & Rescue vessels. Optimization of the superstructure of ships for reduced resistance to still air and to wind has traditionally not been an important subject. Again, there are barriers, in the form of requirements for and the usage of covered spaces. However, for ships with large superstructures and for ships operating at relatively high speeds, there will be a potential for reduction of power consumption by carrying out systematic streamlining of the superstructure to the greatest possible extent. For these ships, it is estimated that there is potential for reduction in power consumption of 2-5%, depending on the size of the superstructure and the area in which the ship operates. Also, for other ships, there is expected to be a certain potential for reduction in power consumption, perhaps in the order of 1-2%, by keeping the topsides as uncluttered and streamlined as possible. The efforts to achieve reductions may range from the simple (such as grinding weld beads flat) to the more extensive (for example, redesigning and repositioning cranes, applying spoilers to alter the airflow over the funnel and deck-houses, and designing more streamlined deck-houses). In the case of transition to electric population system which will be able to transfer engine room to bow-side, streamlining superstructure will be adopted. Integrated streamline hull and superstructure form for PCC PCC has very tall and large ship s main body above water surface from fore-end to aft because of the purpose of this ship type. This is why wind resistance is large, and fuel consumption due to wind resistance becomes very large particularly under rough sea conditions. ratio (%) Approximately 0-2% reduction of power consumption, which corresponds to about 5-30% reduction of wind resistance. For some type of ships, 3% reduction of the power consumption has been achieved. ratio when using with other energy saving (%) This can be combined with all without loss of improvement ratio with the The year expected for in-use of that technology Available by 2020 (partly already achieved. Integrated streamline hull and superstructure exception of some of the wind power options 6(3) form including rear part of ship would be available in 2020s.) Current status of application This technology is normally used for ships with large superstructure. This technology is not suitable for ship types with relatively small superstructure to improve their energy efficiency because of area requirement for superstructure and lower cost benefit. Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) Sometimes It is 0 to hundreds of used with questionable if it thousands of specific ship will make a dollars types, e.g. car difference in carrier, high streamlined speed crafts. design of superstructure for a container vessel as it will often be completely covered by container stacks. A spoiler design on bow (cover) might be helpful on smaller ships. Operational you gain by better distribution of containers. However for tankers and bulkers it will be possible as we also se it for car carriers

135 MEPC 71/INF.12 Annex 4, Page 5 Categorization / Type of Short-description Integrated streamline hull and superstructure form could reduce wind resistance when receiving head wind and could assist fore thrust when receiving bow wind, and this could reduce fuel consumption. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) (3) Optimization of propeller The main abating effect of optimization of the propeller is obtained by increasing the diameter of the propeller and reducing the number of its revolutions per minute (RPM). The requirements to maintain adequate clearances between the propeller and the hull and to attain sufficient submersion of the propeller when the ship is operating in a seaway and/or in ballast condition set restrictions on the extent to which the diameter of a propeller can be increased. A propeller that is operating at a low number of revolutions per minute may require the additional cost of installing a reduction gear, while propellers operating at a higher number of revolutions per minute can generally be directly connected to the main engine. Propellers with a large diameter, operating at a low number of revolutions per minute, will therefore be best suited to deep-draught ships; this includes most tankers and bulk carriers and many general cargo vessels. Such propellers will be less suited to many container vessels, and they will not, in general, be suited to RoRo cargo, RoPax vessels or cruise vessels. Existing ships of the displacement type mostly operate within a moderate Froude number interval (Fn 0.3). This fact dictates the choice of propulsive devices (propellers) that may be used in order to achieve ship thrust. The figure below shows the efficiency as function of the propeller loading coefficient CTh, which is from "Hydrodynamics of Ship Propellers" by Andersen and 0-3% (The Contracted Loaded Tip (CLT) propeller shows the higher improvement ratio, which depends on the ship type, the propeller loading coefficient etc.) Ratio is updated based on the actual performance measured. This could interfere with other energy saving devices especially for devices fitted on stern part of ship. can be combined with most of the other without loss of improvement ratio, in particular categories 1, 2, 4, 5 and 6. The improvement potential of category 7(1) might be slightly reduced. Combination with category 3 Already available Normally used 4 5 % seems high 1-2 % seems more realistic depending on the propeller diameter On 2-stroke engine is it not relevant with reduction gear. We have not seen high savings for new propeller types like CLT or Kappel. In a few cases, we have seen 1-2% saving compared to a traditional propeller design with latest technology. However, both CLT and Kappel often 0 to hundreds of thousands of dollars (The cost depends on the size of engine and propeller.)

136 MEPC 71/INF.12 Annex 4, Page 6 Categorization / Type of Short-description Breslin (Cambridge University Press 1994). The thrust loading coefficient CTh is defined as follows: T C Th = and C 1 2 ρ A 2 Th = 8 R disk V A π C Th = 8 K T A π J 2 disk = π D 2 4 prop J = V A n D K T = R (1 t) ρ n 2 D prop (1 t) ρ (V A D prop ) 2 4 R = (1 t) T V A = (1 w) V where: Dprop = propeller diameter V = ship speed J = propeller advance ratio R = ship resistance T = propeller thrust t = thrust deduction fraction w = wake fraction = density of sea water ratio (%) ratio when using with other energy saving (%) is not applicable. The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology suffer from Cavitation. Many discussions on scaling, but even full scale test has not showed premium performance. Increase of diameter (low rpm) are a more robust solution. This will lower Cth and reduce benefit for propellers designed for high loading. Large propeller diameters can result in problems with cavitation Estimated whole cost *1 (US dollar) Competing goals, such as noise reduction or erosion-free long-term operation, might reduce

137 MEPC 71/INF.12 Annex 4, Page 7 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology the reduction potential Estimated whole cost *1 (US dollar) Figure Efficiency of propulsive devices A weaker but nevertheless considerable lever to improve propeller efficiency is the concentration of its radial load distribution close to the blade tips. This measure is limited by the concurrent increase of vibration excitation but is frequently used in case of retrofit propellers in times of slow steaming. The most efficient type of propeller is a well-designed fixed-pitch (helical) propeller. However, for other reasons, alternative propulsion devices need to be considered. For instance, controllable-pitch (CP) propellers, although slightly less efficient then fixed-pitched propellers, may be

138 MEPC 71/INF.12 Annex 4, Page 8 Categorization / Type of Short-description selected if the ship in question needs to satisfy the requirements of rapidly reversing thrust or efficient operation in significantly different environmental conditions. On the other hand, for ships with demands for high manoeuvrability, propellers with a vertical axis or CP propellers represent a preferable choice. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) There are two examples in relation to the technology for the optimization of propeller, namely, Kappel propeller and CLT propeller. Example 1 Kappel propeller Tip vortices are formed due to the difference in pressure between the pressure and suction side of the propeller as the water will move from the region of high pressure to the region of low pressure. The pressure on both sides near the tip will therefore equalize and the efficiency of the tip region will decrease. The Kappel propeller minimizes the flow over the tip, and the outer region of the Kappel propeller therefore retains a high efficiency increasing 4-5% of the total efficiency of the Kappel propeller compared to conventional propellers. APPEL_Propeller Example 2 CLT (Contacted Loaded Tip) propeller CLT propellers are characterized by the following: The blade tip generates a substantial thrust. The pitch increases from the root to the tip of the blades. The chord at the tip is finite. End plates are fitted at the blade tips, toward pressure side; they are adapted to the fluid vein contraction to reduce as much as possible their viscous resistance.

139 MEPC 71/INF.12 Annex 4, Page 9 Categorization / Type of Short-description The end plates operate as a barrier, avoiding the communication of water between the pressure and the suction side of the blades, allowing to establish a finite load at the tip of the blades. The fundamental goal of the CLT propeller is to improve the propeller open water efficiency by reducing the hydrodynamic pitch angle through the reduction of the magnitudes of induced velocities at the propeller disk. The pressure drop on the suction side of CLT propellers is less than for the conventional equivalent and therefore the extent of the cavitation developed on the suction side is lower and hence the pressure forces that a CLT propeller exerts on the stern hull structure are lower. Additionally, CLT propellers have a reduced tip vortex because of the existence of the end plates. The combination of these circumstances means that the pressure forces exerted by a CLT propeller on the stern structure are of a lower magnitude than for conventional propellers, and so in turn the induced hull vibration and noise levels on board are lower.the noise radiated to the water is also lower for CLT propellers. As the blade area of the CLT propeller is more efficiently used for supplying thrust, the optimum diameter is lower than for an equivalent conventional propeller. CLT propellers offer higher efficiency which may be used to achieve fuel savings at constant ship speed or alternatively ship speed increase at constant fuel consumption. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar)

140 MEPC 71/INF.12 Annex 4, Page 10 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) (4) Podded (azipod) drives Podded (azipod) drives are systems where an electric motor with a propeller is suspended under the appropriate (usually the aft) section of the hull. The pod can be rotated to direct the thrust, resulting in very good ship manoeuvrability. In many cases, more than one pod is used. New pod drives have pulling propellers that face forward. This gives the pod a good flow of water into the pod, resulting in high propulsion efficiency. However, the pod in itself increases the drag, thus reducing total efficiency. Experience from tests of hulls in the towing tank at MARINTEK clearly indicates that the net effect of podded propulsion on the energy efficiency of propulsion is generally negative when compared to conventional designs of propulsion systems. Similar to the effect of conventional propeller. The improvement potential of category 7(1) might be slightly reduced. This can be combined with most of the other without loss of improvement ratio, in particular categories 1, 2, 4, 5 and 6. In use Rarely used Competing goals, such as This could be noise reduction, effective only might reduce for ships which the reduction change the potential course frequently such as tugs, supply vessels and other vessels for short sea shipping. It is not effective for other ship types. Normally used with ship types requiring

141 Categorization / Type of (5) Matching and integration of hull form and drivetrain Short-description An integral approach, system integration, of hullform and drivetrain (propeller, gearbox, main engine) could lead to higher optimisations of solely improvement on a single system. The complete performance of the hull-propeller interaction, driven by the engines, could be assessed in a way where all expected operational conditions are considered. When the whole system is seen as systems which interact and interfere with each other, the optimal solution can be found. Experience from Conoship International shows that proper integration of a hull form (with the ConoDuctTail), nozzle, propeller, gearbox and main engine leads to savings of abt. 20%, with even better results in seaways. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology 1-20% Already available Current status of application superior manoeuvrability Future prospect of the applications / adoptions of the technology MEPC 71/INF.12 Annex 4, Page 11 Estimated whole cost *1 (US dollar) 2 Recovery of propeller energy A considerable number of devices have been invented for improving the power consumption of ships by recovering as much as possible of this rotational energy in the flow from the propeller, or to provide some pre-rotation of the inflow into the propeller. The most important of these will be considered here.

142 MEPC 71/INF.12 Annex 4, Page 12 (1) Coaxial contra-rotating propeller The coaxial contra-rotating propeller is an obvious device for recovering some of the rotational energy. To avoid problems with cavitation, the aft propeller usually has a smaller diameter than the front propeller. The aft propeller is therefore not working on the complete rotating flow field from the forward propeller. In addition, the more complicated shafting results in mechanical losses that offset some of the gain that is obtained by recovering the rotational energy. It is also reported that gearboxes for contra-rotating propellers may present problems. Reported gains in power consumption range from 6% to 20%. Gains of 15% and 16% have been reported from two different full-scale measurements. The improve ratio is affected by the propeller loading (thrust per unit area), different ratios of the diameters of the two propellers, or some other factor(s), and cannot be achieved for every ship. Contra-rotating propeller arrangements require a short shaft line and are therefore primarily suited to singlescrew ships. The arrangement is particularly beneficial for relatively heavily loaded propellers, and the best results (in the form of power consumption) have been found in fast cargo vessels, ro ro vessels and container vessels. Naturally, this type of technology is tested in cases where it is expected to be particularly suitable. The contra rotating propellers have also been used for podded propulsors, where the same efficiency gain of 5 10 % has been obtained. 8% This technology collects the energy of rotational flow. This is why this could interfere with other energy saving devices which have same effect such as pre-swirl devices and post-swirl devices. can be combined with most of the other without loss of improvement ratio, in particular categories 1(1)(2), 4, 5 and 6(1)(2)(3). The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. Already available Sometimes used Application is limited due to its expensive cost. Mostly relevant for highly loaded propellers, e.g. for vessels with draught restrictions. Complicated and expensive installation and maintenance. Expected savings to be in range 0-5% Future application would be limited because of its application cost. Millions of dollars

143 MEPC 71/INF.12 Annex 4, Page 13 Categorization / Type of (2) Free rotating vane wheel Short-description The vane wheel (Grim wheel) is a freely rotating propeller, installed behind the main propeller. The vane wheel has a larger diameter than the main propeller. The part that is directly behind the main propeller is turned by the swirl from that propeller and acts like a turbine, driving the part of the vane wheel that is outside the diameter of the main propeller. This outer part acts as a propeller. Losses in the axially accelerated flow behind the propeller and part of the rotational energy is thus transformed into propulsive energy. s in power consumption are reported to be around 10%. It is claimed that an important benefit of the vane wheel is that the smaller main propeller that can then be installed results in a lighter and less costly propulsion unit. The long and slender vane wings may be damaged at sea, especially in a heavy seaway. It should be noted that, if there is space in the after body for a vane wheel, there will also be space for a main propeller of larger diameter, offering approximately the same improvement in power consumption as the combination of a small main propeller and a vane wheel. The vane wheel should be a suitable potential improvement for cargo ships. ratio (%) 0-10% of power consumption 10 percent would be too large. Therefore, 10 percent should be considered as maximum improvement ratio. ratio when using with other energy saving (%) Will not work together with other post-swirl devices like PBCF, Costa Bulb, Asymmetric rudder This technology collect the energy of rotational flow. This is why this could interfere with other energy saving devices which have same effect such as pre-swirl devices and post-swirl devices. To fit large diameter propeller could cause negative affect to the improvement of hull form. The year expected for in-use of that technology Available Current status of application Rarely used This device is not applied recently loading because there could be a damage to this device under rough sea. Future prospect of the applications / adoptions of the technology No experience. Again, will saving be related to of propeller. With large diameter propeller, this is not found to become relevant Competing goals, such as noise reduction, might reduce the reduction potential Estimated whole cost *1 (US dollar) This can be combined with

144 MEPC 71/INF.12 Annex 4, Page 14 Categorization / Type of (3) Ducted propeller Short-description The ducted propeller consists of a propeller mounted centrally in a ring foil. Compared to the conventional propeller of the same diameter and thrust, this arrangement allows a larger mass of water to be supplied to the propeller, improving the operating conditions around the propeller and the ideal efficiency. The duct generates additional thrust. The potential for reduced power consumption on relevant ships has been reported to be in the range 5-20%, with perhaps 10% being a good average value. The duct results in increased resistance, but at higher propeller loadings this is more than compensated for by the positive effects of the combination ratio (%) 0-10% of power consumption The improvement ratio depends on the thrust loading condition. Only useful for ship types with ratio when using with other energy saving (%) most of the other without loss of improvement ratio, in particular categories 1(1)(2), 4, 5 and 6(1)(2)(3). The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable This can be combined with most of the other without loss of improvement ratio, in particular categories 1(1)(2), 4, 5 and 6(1)(2)(3). The year expected for in-use of that technology Current status of application Available Rarely sometime used Not relevant for container vessels This device could be effective for ships with high propeller load. Future prospect of the applications / adoptions of the technology 10 % seems a little high however good experiences on tug boats Estimated whole cost *1 (US dollar)

145 Propeller efficiency (-) MEPC 71/INF.12 Annex 4, Page 15 Categorization / Type of Short-description of propeller and duct. Below is the diagram that illustrates how the propeller efficiency depends on the thrust loading coefficient CTh (defined in 1. (3)) Ideal efficiency Conv. propeller (Wageningen B-series) Ducted propeller Propeller efficiency ratio (%) high propeller loading. ratio when using with other energy saving (%) The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. The year expected for in-use of that technology Current status of application This has not been applied for conventional cargo ships. Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) Propeller thrust loading coefficient Cth (-) The ideal efficiency is a purely theoretical efficiency assuming no losses. The curves are based on the previous diagram by Andersen and Breslin and show that only for ships with a CTh value higher than 2 ducted propellers can be recommended. It could also be mentioned that some propeller manufacturers have improved the efficiency of ducted propellers by further optimization of the duct, such that the propeller thrust at low speed (bollard pull condition) has been increased by 6 8 %. Ducted propellers are therefore suited for ships operating at high propeller loadings, such as tankers, bulk carriers, tugs and different offshore supply and service vessels. (4) Pre-swirl devices These are devices that aim to provide a favourable prerotation of the flow of water in front of the propeller. They 1-4% improvement of This could interfere with Available Sometimes - normally used Tens of thousands to

146 MEPC 71/INF.12 Annex 4, Page 16 Categorization / Type of Short-description include radial reaction fins in front of the propeller and an asymmetric stern. Considering radial reaction fins, a reduction in power consumption has been given as 3-8% from tests with models, while the result that has been reported from a full-scale test was 7-8%. For the asymmetric stern, improvements in power consumption of 1-9% have been reported in tests on models. The improve ratio differs according to the ship type or the propeller characteristics and so on. Radial reaction fins or an asymmetric stern should be applicable to all single-screw ships, and should work according to expectations in many cases. It should be noted, however, that in many cases the expected benefits have not been demonstrated in fullscale operation. Also, there are two examples in relation to Pre-swirl devices, namely, Guide vanes in front of the propeller and Wake-equalizing duct should be integrated in-use. Example1 Guide vanes in front of the propeller The aim of guide vanes is to eliminate or reduce the crossflow that is often observed in front of the propeller. These vanes are fitted in front of the propeller on both sides of the sternpost. The vanes straighten the flow in the boundary layer in front of the propeller, thereby improving its efficiency. Cross-flow appears mostly in ships with stern bulbs and full hull forms that operate at relatively low speed. The benefit is therefore largest for tankers and bulk carriers. The improvement decreases with decreasing fullness of the hull form. Example2 Wake-equalizing duct The wake-equalizing duct consists of one half-ring duct with foil-type sections attached on each side of the after ratio (%) the power consumption ( 4-6% improvement would be achievable in case post-swirl device in 2.(5)) is used together) The improvement ratio is reduced because of interfere of devices. Full scale tests on an optimised PSS indicate 7% improvement (PD) for a bulk carrier (ref. EU GRIP project, ncedirect.com/s cience/article/pi i/s ) ratio when using with other energy saving (%) CRP or other pre-swirl devices. The effect of this technology depends on stern hull form. This device collects stern viscous vortex, and this is why the effect is larger for fat ships than slender ships. 8% improvement would be achievable in case post-swirl device in 2.(5)) is used together can be combined with most of the other without loss of improvement ratio, in particular categories The year expected for in-use of that technology Current status of application Our experience with PSS (DSME) and with BTF (Becker) are predicted saving in range 2-4%. We do not have any full scale PSS installation, but other owners have experienced structural problems. We have no full scale performance records for BTF. Previous experience with Mewis duct on VLSS showed a saving of 4-5% which was confirmed in full scale. Mewis ducts are not suitable for Container vessels. Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) hundreds of thousands of dollars

147 MEPC 71/INF.12 Annex 4, Page 17 Categorization / Type of Short-description body, in front of the propeller. The half-ring duct accelerates the flow into the propeller in the upper quadrant on each side and retards the flow in the lower quadrants. This results in a more homogeneous wake field in front of the propeller, while the average wake is almost unaltered. The improved power consumption that is obtained from well-designed wake-equalizing ducts results from several component savings: improved efficiency because of more axial flow and a more homogeneous wake field; reduced resistance because of reduced flow separation at the after body; lift on the ducts directed forward; orientation of duct axes so that the inflow to the propeller is given a small pre-rotation; and improved steering, due to straightened flow over the rudder and more lateral area aft. ratio (%) ratio when using with other energy saving (%) 1(1)(2), 4, 5 and 6(1)(2)(3). The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. The year expected for in-use of that technology Current status of application This is not used for high-speed ships. Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) (5) Post-swirl devices A number of devices belong to this category. Several of them involve modifications to the rudder. The most important among these devices may be additional thrusting fins at the rudder, rudder bulb systems with fins, fins on the propeller fairwater (boss cap fins) and an asymmetric rudder. For these devices, improvements in power consumption of 1-8% have been reported from tests on models. From full-scale measurements, an improvement of 8-9% has been measured for additional thrusting fins at the rudder, while 4% has been reported for boss cap fins. Post-swirl devices should be applicable to all new ships, but, as for pre-swirl devices, the benefits, in many cases, have been difficult to demonstrate in fullscale operations. 1-4% improvement of the power consumption ( 4-6% improvement would be achievable in case pre-swirl device in 2(4) is used together) This could interfere with CRP or preswirl devices. The effect of this technology depends on stern hull form. 8% improvement of power consumption would be achievable in case pre-swirl Already available Sometimes - normally used Savings of 2-4% for a welldesigned asymmetrical with rudder bulb have been seen. Requires a propeller cap that closes the gap to rudder bulb. Savings from PBCF between Tens of thousands to hundreds of thousands of dollars

148 MEPC 71/INF.12 Annex 4, Page 18 Categorization / Type of Short-description Also, there is an example in relation to Post-swirl devices, namely, integrated propeller and rudder units should be integrated in-use. Example 1 Integrated propeller and rudder units As the name implies, the propeller and rudder are designed as an integrated unit, part of the design being a bulb behind the propeller that is fitted into the rudder. At least two patented designs exist. The effect of these units has been reasonably well documented in tests on models and in full-scale trials. An improvement of 5% in power consumption may be taken as typical. The units are applicable to general cargo vessels, RoPax vessels and container vessels operating at relatively high speed. ratio (%) ratio when using with other energy saving (%) device in 2(4) is used together This can be combined with most of the other without loss of improvement ratio, in particular categories 1(1)(2), 4, 5 and 6(1)(2)(3). The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. The year expected for in-use of that technology Current status of application 0-2% are found. Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) (6) Split stern (Twin skeg) Split stern has effect on improvement of propeller efficiency caused by mitigation of propeller loading, and is suitable to beamy ship. 0-6% Potential in application highly depends on the This can be combined with most of the other without loss of Already available Rarely sometimes used Biggest gains can be 4 6 % realistic however require a big (and expensive) CAPEX Cost would be significantly increased because of complex stern hull form.

149 MEPC 71/INF.12 Annex 4, Page 19 Categorization / Type of 4 Engine energy recovery (1) for lowspeed engine Short-description Energy-recovery systems for ship engines have been available from producers of low-speed engines for many years. The first system offered was usually based on an additional exhaust turbine which was fed from the exhaust receiver by a fraction of the engine exhaust flow, in the range of 10%. The exhaust turbine could be connected to the engine's crankshaft or alternatively to an electric generator. Since fuel prices in the past have been too low to make these systems profitable, the number of installed systems is relatively small. For some years, more advanced systems have been developed and are today commercial, at least for lowspeed engines. An example is B&W's system TES (thermo efficiency system), which combines a turbine in the exhaust gas with a steam cycle that is driven by exhaust heat and running a steam turbine. The two ratio (%) application case. Waste Heat Recovery System could recover 2-6% reduction of power consumption. 6-12% seems high 3-6% is more realistic The effect of this device could be taken into account as reduction of ratio when using with other energy saving (%) improvement ratio, in particular categories 1, 4, 5, 6 and 7. The improvement potential of combinations within category 2 and with 1(1), 7(1) might be slightly reduced. This can be combined with almost all of the other without loss of improvement ratio. Combination with category 4(2)(4) is not applicable. The year expected for in-use of that technology Already available Current status of application achieved if right combination of engine can be installed in order to have a large propeller diameter as possible. This could be applied only for ultra-wide ships. This is not suitable for conventional ships. Sometimes normally used This device is applied for larger ships which are suitable for this device. Future prospect of the applications / adoptions of the technology Effective for ships with low depth and high power Application comes with the risk of poor wakefield and related problems with cavitation, vibrations, shaft bearings. Efficient combination with EGCS Estimated whole cost *1 (US dollar) Hundreds of thousands to millions of dollars

150 MEPC 71/INF.12 Annex 4, Page 20 Categorization / Type of (2) for mediumspeed engines Short-description turbines are coupled to a generator for production of electrical power. The power can then be used to drive a shaft generator/motor to assist the main engine, or consumed elsewhere in the ship. The corresponding increase in engine power is estimated to be in the range of 9 to 11%, which, in terms of shaft efficiency, increases to about 55% (from about 49.5%). The contributions from the two systems are respectively 5% and 6%, from the exhaust turbine and the steam turbine. The efficiency of the steam cycle is somewhat limited by the minimum recommended temperature of the exhaust stack, which must be above 180 C to control the formation of deposits and the corrosion by sulphur oxide that are related to the use of fuel oils. Steam cycles, as a means of energy recovery, have some properties that are quite challenging on board a ship. The relatively low temperature level makes systems relatively bulky. In particular, the condenser operates at the low steam density that exists at the actual condensation temperature. There are some interesting forthcoming developments that are expected to make a significant impact on the gain in engine efficiency. Organic Rankine Cycle systems have been designed and are already commercial. They show some favourable properties, in particular much smaller space requirements compared to a steam system. The working fluid is currently alkanes or refrigeration fluids. Due to the properties of the working fluids (fire hazard, ozone-depletion properties), high-pressure CO2 is considered to be a more desirable working fluid. The medium-speed diesel engines have a lower fuel efficiency compared to low speed ones, usually in the range 42% to 44.5%. These engines normally have an ratio (%) PAE. Maximum effect of this would be about 5%. The effect of this technology would be smaller if temperature of exhaust gas is relatively low. 3.5% of shaft power 5% by turbo Generator; and 10% by turbo Generator + Power Turbine 3,5 % seems high The effect of this device could be taken into account as reduction of PAE. Maximum effect of this would be about 5%. ratio when using with other energy saving (%) This can be combined with almost all of the other without loss of improvement ratio. Combination with category 4(1) is not applicable.. The year expected for in-use of that technology Available Current status of application Sometimes used This technology is applied for specific ship type. To fit this would be difficult because of its arrangement. Future prospect of the applications / adoptions of the technology Efficient combination with EGCS Estimated whole cost *1 (US dollar) Millions of dollar

151 MEPC 71/INF.12 Annex 4, Page 21 Categorization / Type of Short-description exhaust gas temperature in the range of 300 C to 360 C. While the minimum temperature of the exhaust stack is 180 C or above, which poses a limitation to heat recovery, the energy utilization is calculated to be in the range of about 3.5% of the shaft power. ratio (%) The effect of this would be smaller if temperature of exhaust gas is relatively low. ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar)

152 MEPC 71/INF.12 Annex 4, Page 22 Categorization / Type of (3) for gasfuelled engines Short-description Due to increasing prices of fuel oil and regulations to control exhaust emissions (restricting operations or taxation), there is an increasing interest in using gas engines, burning natural gas, in marine applications. The currently available gas-fuelled engines for ship propulsion, with piston bores in the range of 25 cm to 50 cm, have slightly higher shaft efficiency (at MCR) compared to their diesel counterparts, normally in the range of 44.5% to 47% depending on engine size and engine concept (sparkignited and dual-fuel). The gas engines offer a higher potential for energy recovery. This comes from the higher exhaust temperatures (normally in the range of 400 C to 430 C) and a possibility to run with an exhaust stack temperature below 100 C. This can be done, while the fuel does not contain any sulphur and the combustion produces very small quantities of particles. Thereby, the exhaust system is likely to be little affected by the low temperature of the exhaust. ratio (%) Reduction of CO2 22% (including 13% energy shaft power recovery) 5% by turbo Generator; and 10% by turbo Generator + Power Turbine ratio when using with other energy saving (%) This can be combined with almost all of the other without loss of improvement ratio. Combination with category 4(1)(2) is not applicable.. The year expected for in-use of that technology Current status of application Available Rarely - sometimes used Future prospect of the applications / adoptions of the technology Infrastructure for LNG fuel should also be provided. There are few infrastructure available in the world at this time. Estimated whole cost *1 (US dollar) More than millions of dollars A simplified calculation, based on an exhaust gas temperature of 430 C and an exhaust stack temperature of 50 C, gives a theoretical (Carnot) efficiency of about 32.5%. Considering a turbine efficiency of 80% and some additional losses related to pumping and heat exchangers, the total recovery efficiency could be in the level of 22%. Using actual figures for exhaust heat flow and engine shaft power, the energy-recovery figure is 13% of the shaft power. That means that the actual shaft efficiency increases from 45% to 50.9%. (4) Machinery arrangement <Comment: Clarification is necessary what kind of gases are taken into account in this explanation.> Currently, tankers, bulk carriers, containerships and CRP-Pod general cargo ships have one large low-speed propulsion vessel could The effect of this would Already achieved Rarely used

153 MEPC 71/INF.12 Annex 4, Page 23 Categorization / Type of and hybrid propulsion Short-description engine directly connected to the propeller. This arrangement has proven to be very efficient and, since the ships operate mainly at high engine load, there is little to gain by complex multi-engine machinery arrangements or by using hybrid propulsion systems. For the RoPax/cruise segment, it is currently common to use multiple engines and two or more propellers. A primary reason for this is restrictions on draught and high power demand. An additional reason is the space restrictions and the use of medium-speed engines; hence there is already a need (and transmission loss) for a gearbox. This multi-engine situation opens up some possibilities for designing systems that can handle variable loads. There are some applied for certain types of ships. Today most super yachts have geared diesel driven twin screw propulsion with open shaft lines and twin spade rudders. However a trend towards hybrid propulsion has emerged. The main reason for this is the varied operating profile of super yachts and the consequences thereof for propulsion efficiency. ratio (%) improve the propulsion performance by 13% compared with the conventional twin shafts and propellers system. The effect of this technology would be lower. ratio when using with other energy saving (%) interfere with pre-swirl and post-swirl device can be combined with almost all of the other without loss of improvement ratio. Combination with category 4(1) is not applicable. The year expected for in-use of that technology Current status of application Normally used with suitable ship types In general, CRP is more effective than CRP-Pod. Thus this technology is only applied for high-speed ROPAX and other limited ship type. Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) A recent survey of a sample of 27 yachts has shown that over 50% of the operating time was at less than 20% of engine load. Less than 5% of the operating time was over 80% engine load. Annual average operating hours were as low as 277 per yacht (MEPC 65-INF.15). Such operating profiles differ considerably from those of most merchant vessels. Hybrid propulsion systems enable higher propulsion efficiencies at all engine loads.

154 MEPC 71/INF.12 Annex 4, Page 24 Categorization / Type of Short-description <Comment: this goes beyond the short description of technology; contains information that could be used when considering the potential of combined > ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) Information for the Hybrid propulsion (diesel-electric propulsion combined with use of batteries for energy storage): Most diesel engines have their optimum performance (lowest specific fuel oil consumption, SFOC) at % MCR (if not specially tuned for low load operation where the lowest SFOC occurs at a lower MCR percentage). Some diesel-electric propulsion systems operates at a variable load situation which means that multiple gen-sets occasionally operates at very low load, i.e. far away from the optimum point resulting in a high SFOC and sometimes at a very high load close to 100 % MCR. For such systems it is beneficial to change the operation profile such that all gen-sets are operated at the optimum point, such that the surplus of electric power is accumulated in batteries for use in the periods where there is a need for high propulsion power. With such a socalled hybrid system often one or two gen-sets can be omitted, as the extra electrical is taken directly from the battery package. Hybrid systems are comfortable on diesel-electric ferries on short routes where there is a very high power demand during the acceleration phase, while the power demand is limited when the service speed has been achieved. (5) Fuel-cell propulsion (Revised by CESA) Fuel cells have high potential thermal efficiency and low emissions. For this reason, fuel-cell technology is, in principle, also an interesting alternative to the use of traditional combustion engines for merchant shipping. Fuel cells can be used either as standalone or in a (5-10%)* 25-30% (100%) * aux. engines only This can be combined with all other without loss of Under development 2020* Never used for conventional commercial vessel Today only systems with small power output available

155 MEPC 71/INF.12 Annex 4, Page 25 Categorization / Type of Short-description combined cycle, where exhaust heat is recovered for additional generation of power. Fuel-cell systems have been identified as particularly promising power generators for both ship hotel power and also for hybrid propulsion systems, where they work in combination with a diesel engine. There are many issues relating to the use of fuel cells on board ships. Fuel cells use non-conventional fuels, such as hydrogen, methanol, low-flashpoint diesel, some requiring fuel preparation (reforming). With decreasing price and increasing reliability, fuel cells presently become more interesting than before. Main technological obstacles to operating fuel cells on board medium to large ships have been removed and international safety standards are in process of finalization. The related R & D projects have reached the implementation stage with full scale on-board trials, including: development of fuel processing systems for fuel-cell units capable of running on liquid fuels; energy-recovery systems (e.g., boilers, turbines) for use in conjunction/integration with high-temperature fuel-cell systems (MCFC and SOFC); standardization of fuel-cell systems (including auxiliary systems) into modules of 0.5 MW to 1.0 MW size; intrinsically safe systems for onboard storage of fuel and fuel handling; and development and full-scale validation of systems with respect to their use in the marine environment: reliability, availability, vibration, accelerations, salinity, humidity, and ability to respond to transient power demands. ratio (%) in short sea shipping with hydrogen produced with renewable energy ratio when using with other energy saving (%) improvement ratio.. The year expected for in-use of that technology Current status of application Trial application on-board ferries, container ships and offshore vessels, Cruise ships in 2018/2020 Future prospect of the applications / adoptions of the technology There would be difficulty of application because of cost and size. Timely finalization of IMO regulation and availability of appropriate FC primary fuels, such as methanol, lowflashpoint diesel or hydrogen Estimated whole cost *1 (US dollar)

156 MEPC 71/INF.12 Annex 4, Page 26 Categorization / Type of (6) Boiler efficiency with Waste Heat Recovery (Economizer) Boiler configurati on Design with burner having VFD plus damper Adding a pre- heater (waster heat recovery in conjunction with a boiler (7) Optimization of Exhaust gas via Waste Heat Recovery ( Main & Aux engine) Short-description In cogeneration (Combined Heat and Power, CHP) applications, a conventional power plant producing electricity is enhanced with a heat recovery system to utilize the energy otherwise wasted in the exhaust gas. Since the exhaust gas waste heat is a free source of energy; the more you recover - the more you save. Replacement of the damper air control with a Variable Frequency Drive can improve the fuel usage for the ship. VFD provides electrical savings that can be due to oversize fans, oversize motors, and minimizing damper horse power due to lower damper pressure drop. Energy saving could vary between 5 to 30% depending on boiler firing rate. A Waste Heat Recovery Unit can be added to the boiler exhaust so that latent heat exhaust can be used to preheat the feed water. Increasing heating efficiency by reducing the amount of energy needed for heating the water to create steam. Additionally, when the WHR is used together with a condenser the flue gases are cooled down to condense water vapour from burning of hydrogen, allowing extra energy recovery. The WHR coupled with boiler will increase boiler efficiency by a further 10% to 15%. Waste heat recovery systems recover the thermal energy from the exhaust gas and convert it into electrical energy, while the residual heat can further be used for ship services (such as hot water and steam). The system can consist of an exhaust gas boiler (or combined with oil fired boiler), a power turbine and/or a steam turbine with alternator. Redesigning the ship layout can efficiently accommodate the boilers on the ship to better fit these systems. In engine systems much of the energy produced by the engines goes to heat rather than power - 25% of an engine s fuel energy goes into the exhaust flow as waste ratio (%) 30% (VFD) more efficient at a firing rate of 80% ratio when using with other energy saving (%) WHR plus VFD = 5% increase in efficiency (30% +5%) 3.9 to 5.6% Up to 15% when used in conjunction with a condenser The year expected for in-use of that technology Already available (VFD not chosen because of higher cost) Already available Current status of application Sometimes used Normally Used Main Engine Waste Heat Recovery Sometimes used for auxiliary engine Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) Cost for VFD is between US$ to US$ excluding installation. Waste heat recovery standalone 25,000.00US plus 10,000US for the electric generator

157 MEPC 71/INF.12 Annex 4, Page 27 Categorization / Type of Short-description heat. By recovering this energy with a boiler or waste heat recovery system (WHR) it is possible to reduce both fuel consumption and emissions. The converted energy can be used for production of steam and hot water, production of extra electricity, heating of fuel oil and/or supplying district heating and cooling. The waste heat from the auxiliary engines has not been considered in the past, but it contains a large amount of energy which can be utilized to supplement steam requirements during port stays and - for some vessels - also during voyage. Unlike the continual operation of the main engine during oceangoing voyages, the operation of the auxiliary engines varies. The WHR, therefore, has been developed as a customized solution focused on generating energy under varying load conditions. To ensure the most advantageous design, the Aalborg WHR will be specially tailored to the individual ship and engine design with due consideration to the existing uptake back pressure and other critical factors The WHR concept has been developed as a customized solution with special focus on energy generation compared to return on investment and has a very short payback time of down to 4-6 months in optimum cases, but normal payback time will be approximately 1 to 1½ year depending on the number of days the produced steam can be utilized (offset against the steam requirement from the oil-fired boiler) and the redundancy requirements. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) When the WHR is used together with a condenser the latent heat of the water vapor is recovered. The lower the water temperature, the greater the amount of energy recovered.

158 MEPC 71/INF.12 Annex 4, Page 28 Categorization / Type of Short-description Overall, Lowering the fuel oil consumption on the main engine, through increased efficiency therefore results in higher fuel oil consumption of the oil-fired boiler if the entire steam boiler plant remains unchanged ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) (8) Exhaust aftertreatment solutions (SCR) SCR-Technology can be used for Tier II and for Tier III NOx applications. Especially when operated in conjunction with a Tier II requirement, the NOx reducing performance of a SCR-system can be utilized in a way that the engine is to be optimized to the best possible specific fuel consumption (SFC), accepting elevated engine NOx out emissions, which are reduced to the required Tier level by use of a SCR-system. Optimized SFC results in reduced CO2-emissions. 3-5% This can be combined with other Already available Already increasingly used for Tier III NOx compliance; Rarely used for EEDI improvement Application as Tier III NOx compliance only for new ships slow market roll out CAPEX 0% when used with an engine for Tier III application 3-10% of engine costs when specifically mounted for EEDI improvement OPEX Impact to OPEX depend on correlation of fuel price and Urea price Due to improved SFC, a range between cost neutrality and up to 3% OPEX reduction is possible

159 MEPC 71/INF.12 Annex 4, Page 29 Categorization / Type of 5 Hull coatings (1)Selection of coatings (2)Polymers and air lubrication Short-description In a research project managed by MARINTEK, a number of coatings, of both anti-fouling and foul-release type, from five major manufactures have been tested on 16 Norwegian ships over a period of seven years. The results show that these new coatings are equally as effective as TBT-based systems. (TBT-containing paints are banned as a result of the International Convention on the Control of Harmful Anti-fouling Substances on Ships, 2001 established by IMO.) Frictional resistance can be reduced by modifying the wetted surface of the hull, such as by introducing riblets that mimic shark scales or by applying an artificial enhancement (such as the use of air bubbles and/or air cavities and polymers). Research is still going on concerning air lubrication on hull forms for conventional ships, but it has so far not provided significant improvements. Air-lubrication technology is claimed to provide reductions in resistance that are in excess of 5%, which is significant in this context. ratio (%) 1-2% reductions of frictional resistance by Low friction coating The effect of this could be taken into account as the difference of Vref at sea trial. This is why the effect to attained EEDI is only 1-2 %. 1-3% (The improvement ratio is different according to the ship type; higher improvement ratio could be expected for a blunt ship such as a tanker and ratio when using with other energy saving (%) The effect would be reduced when using other friction reducing devices This can be combined with all other without loss of improvement ratio. The improvement potential of combination with 5(2) might be slightly reduced. The effect would be reduced when using other friction reducing devices can be combined with most of the other The year expected for in-use of that technology Already achieved. Already achieved. Current status of application Sometimes - normally used Hardly a design feature. But sure right coating will reduce fuel. Previous experience with micro bubbles did show a very small net gain. We are looking into updated. Predicted propulsion savings of some 4-5%, but Future prospect of the applications / adoptions of the technology Tightening of environmental policies regarding permissible biocides resulting in lack of efficient coatings Estimated whole cost *1 (US dollar) Tens of thousands to 1.5 million dollars. (There are various type of coating with different cost and effect.) Millions of dollars

160 MEPC 71/INF.12 Annex 4, Page 30 Categorization / Type of Short-description Adding a small amount of polymer to a turbulent Newtonian fluid flow can result in a reduction of the viscous frictional resistance. During the past three decades, numerous research activities were dedicated to the reduction of frictional resistance by applying polymers. As a result, roughly three main methods of friction reduction by polymers have been developed. The first method is based on a molecular scale, due to the fact that the behaviour of polymer molecules in various model flows has been studied. The second type of method relies on investigation of the effects of polymers on the timeaveraged turbulence statistics, while the third type of method examines changes in the coherent turbulent structure due to the presence of polymers. As in the previous case of air-lubrication technology, the three methods of using polymers to reduce frictional resistance are not yet mature, i.e. research in that direction is still going on. Additionally, the concept of continuously injecting polymers into the water may not be suitable for sustainable operation. Therefore, the concept of polymer injection is not considered to be very important for reduction of ship resistance. However, it should be noted that any improvements to the wetted surfaces of the hull that are achieved by these means may also inhibit organic growth. None of the mentioned are proven in service. Additionally, an air-bubble system would require energy to produce the bubbles. ratio (%) a bulk carrier, than a slender ship such as a container carrier.) The effect of air lubrication would be reduced if flat part of bottom is narrow or depth is large. Deeper depth requires more energy for providing air to the bottom. ratio when using with other energy saving (%) without loss of improvement ratio. The improvement potential of combination with 5(1) and 7(1) might be slightly reduced. The year expected for in-use of that technology Current status of application a loss for air bubbles production will eat from savings Surface treatments need to be mechanical robust and have good anti fouling properties such any gain are sustainable. Think expected savings are on high side. Rarely used Based on the current technology, this is only effective to ships with wider and shallow hull. Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) Hull coatings based on nanotechnology have been advertised by different companies for some time now, and have also been mentioned in the media recently. It is claimed that these coatings have the potential of reducing the basic viscous frictional resistance of the underwater hull to a considerable extent and to delay the onset of

161 MEPC 71/INF.12 Annex 4, Page 31 Categorization / Type of Short-description marine growth for an extended period. The claims are largely unsubstantiated at present, but, if they can be even partly realized in the future, power reductions of perhaps 15% may be expected. Thus, this type of coating of the underwater hull will be one of the most important contributions toward reducing fuel consumption and CO2 emissions for well-designed conventional ships. It will be particularly favourable that such coatings probably can be applied both to new ships and to existing ships. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) 6 Alternative fuels and energy (1)Liquefied (Revised by CESA) natural gas Gas that is stored in the liquid state, as liquefied natural gas (LNG), is predicted by many as a forthcoming fuel for ships. Key drivers for this expected development are the low emissions of nitrogen oxides (NOx), SOx and particulate matter (PM) from LNG-fuelled ships. Also, LNG contains more hydrogen and less carbon than diesel fuels; hence emissions of CO2 are reduced. Unfortunately, increased emissions of methane (CH4) reduce the net effect. The price of LNG is significantly less, compared to distillate fuels; therefore there is a considerable economic incentive for a move towards using LNG. The most important technical challenge is finding the necessary space for storage of the fuel on board the ship and the availability of LNG in the bunkering ports. Therefore, LNG is primarily interesting in a regional shipping context, where the ship's range is less of an issue and the next port of bunkering is more predictable. LNG could also become an interesting fuel for tankers, since there is considerable space available for the LNG fuel tanks on deck. LNGfuelled ships would be particularly attractive in NOx emission control areas, since they can meet Tier III 20%-25% This can be combined with all other without loss of improvement. Currently in-use where LNG is available Rarely sometime used Infrastructure for LNG fuel should also be provided. There are few infrastructure available in the world at this time. Concerns methane and infrastructure on slip World-wide implementation is governed by the growth of supply infrastructure and harmonization of safety requirements shore-side Millions to tens of millions of dollars

162 MEPC 71/INF.12 Annex 4, Page 32 Categorization / Type of Short-description emission levels without after treatment of the exhaust gases. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) LNG-fuelled ships can use either pure gas-fuelled engines or dual-fuel engines that are capable of burning gas, diesel or combinations of these. LNG is a proven technical solution, with 10 ships already in operation and 19 ships on order. (2)Biofuels Currently, the cost of bulk LNG is about the same as that of residual (heavy) fuel oil, and significantly cheaper than distillate (fossil) fuels. Natural gas can also be processed to produce Fischer Tropsch diesel, for use in diesel engines; however, in this case, the NOx benefit that is associated with LNG operation would be lost. Also, natural gas can be reformed on site and used as fuel for fuel cells; however, this is currently not considered to be an interesting option due to the principal fuel-cell challenges (including cost, durability and power density). Presently, only four-stroke medium-speed engines for direct-drive LNG propulsion are already on the market. These fuels include current, "first-generation" biofuels made from sugar, starch, vegetable oil or animal fats, using conventional technology. Among these, biodiesel (i.e. Fatty Acid Methyl Esters, FAME) and vegetable oils can readily be used for ship diesel engines. In rough terms, biodiesel could substitute distillate fuels and vegetable oils could substitute residual fuels. With some biofuels, there may be certain issues such as stability during storage, acidity, lack of water-shedding, plugging of fuel filters, wax formation and more which suggest that care must be exercised in selecting the fuel and adapting the engine. Blending bio-derived fuel fractions into diesel or heavy fuel oil is also feasible, from the technical perspective; however, compatibility must be checked, as This can be combined with all other without loss of improvement. Already available Rarely used Availability of fuel These fuels might require adaption of engine technology and/or fuel treatment, handling and storage procedures on board

163 MEPC 71/INF.12 Annex 4, Page 33 Categorization / Type of Short-description is also the case with bunker fuels. Future processes to convert biomass into liquid fuels can be designed to synthesize various fuels that are suitable for use on board ships. Currently, biofuels are significantly more expensive than oil-derived fuels. This would have to change if there is to be an incentive to use these fuels on board ships. Moreover, as discussed in the future scenarios, as long as there is a demand, driven by legislation, for biofuels to be used and for carbon reductions on shore, it will be natural to preferentially use biofuels on land, where this is credited, rather than on ships. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Marine regulations for safety requirements to be adapted and to be finalized Estimated whole cost *1 (US dollar) (3) LPG Liquefied Petrol Gas (ref % This can be combined with most other without loss of improvement ratio (see comments for LNG) Already available Rarely used promising option for LPG tankers; amendment of the IGF Code required (4) Alcohol (Methanol) Same as liquefied natural gas (LNG), methanol is considered as one of alternative fuels which could reduce environmental pollution to the atmosphere and the ocean. Like LNG, it does not contain sulphur, and this could contribute reducing particulate matter (PM) and NOx emissions. Regarding carbon dioxide (CO 2), methane (CH 3 OH) contains much hydrogen, like methane (CH 4) which is the main component of LNG, and this can reduce 10% emissions compared with heavy fuel oil. Methanol has good flammability, but it is difficult to ignite. For dual fuel low-speed diesel engines, Methanol 7-10% Ethanol 3-5% This can be combined with most other without loss of improvement ratio (see comments for LNG) Recently developed Rarely - sometimes used Infrastructure for methanol fuel should also be provided. There are few infrastructure available in the world at this time. promising option for all

164 MEPC 71/INF.12 Annex 4, Page 34 Categorization / Type of Short-description the pilot fuel is injected to ignite and explode methanol fuel. Since a small amount of pilot fuel oil is used, 99% of PM is reduced, and up to 30% of NOx is reduced compared with heavy fuel oil. In comparison with LNG, the LNG fuel system requires ultra-low temperature technology, whereas methanol is liquid at normal temperature, easy to handle such as transportation and storage. Initial cost for methanol is about 1/3 of the cost for LNG. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology ship types because fuel is liquid and can be stowed onboard in integral tanks; amendment of the IGF Code required Estimated whole cost *1 (US dollar) (Methanol) (ref. Effect to CO2- Emissions, Reference to DMX 1 Diesel /Gas Oil (DMX) Light Fuel (LFO) + 1 to + 3 % 3 Heavy Fuel Oil (HFO) + 2 to + 4 % 4.1 Liquefied Petroleum Gas (LPG) Propane 4.2 Liquefied Petroleum Gas (LPG) Butane 5 Liquefied Natural Gas (LNG) - 10 to - 15 % - 10 to - 15 % - 20 to - 25 % 6 Methanol - 7 to - 10 %

165 MEPC 71/INF.12 Annex 4, Page 35 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) 7 Ethanol - 3 to - 5 % The effect of different fuel types to the calculation of the attained EEDI can be ascertained in the following way (ref. MEPC.281(70): The energy content (lower calorific value, LCV) of fuels, as proxy for the specific fuel consumption (SFC) of engines, is correlated to a reference fuel such as DMX (ISO 8217) and multiplied with the CF-value which stands for the carbon content on a mass-basis of the fuel concerned. The results represent the effect {%} of the different fuel types to CO2-emissions in correlation to the reference fuel DMX. (+) elevated CO2 emissions, (-) reduced CO2 emissions. The mechanical efficiency is assumed to be comparable for engine. (5)Wind power/ Auxiliary wind propulsion Wind power can be utilized in various ways on ships. These include:.1 Traditional sails;.2 Solid-wing sails;.3 Kites; and.4 Flettner-type rotors. Although sails were once the only source of propulsion, sails are currently considered to be interesting for providing additional supplementary power, as is suggested by recent studies, for instance. The use of traditional sails will impose bending moments to the hull, causing ships to list. Strength issues could result in a need for masts to run down to the keel, and the presence of the mast and rigging could have significant impacts on cargo handling. Kites differ from other concepts of wind power by having a small footprint during installation and hence being quite feasible to retrofit. Drawbacks with the kite Flettner Rotor type applications in good wind conditions can achieve approximately 5%/unit long term average fuel savings in typical bulker and tanker operations. 5%-15% [100%] This can be combined with some of the other without loss of improvement ratio, in particular categories 5, 6, and 7. The improvements potential of of category 1 and 7(1) might be Already available Sometimes used on cruise ships, ferries, ro-ro ships and yachts Not relevant for container vessels due to lack of deck space. No obstacles for auxiliary use of wind power; sail only propulsion difficult due to safety concerns For Flettner rotor applications 400k-1M/unit when installed onboard, depending on model

166 MEPC 71/INF.12 Annex 4, Page 36 Categorization / Type of Short-description systems include the complex launch, recovery and control systems that are needed. Also, the durability of the lightweight materials that are needed for kite sails is a challenge. Wing sails are solid structures resembling aircraft wings, which provide more thrust with less drag than conventional sails. Flettner-type rotors generate thrust from a rotating object in wind, taking advantage of the so-called Magnus effect. These systems have different characteristics with regards to how the thrust that is generated relates to other parameters, such as wind angle, wind strength, wind stability and ship speed. ratio (%) for ship types with available deck space Based on case studies for a big tanker and bulker Flettner Rotors reduce EEDI by 6-7%. ratio when using with other energy saving (%) slightly reduced. The engine / fuel related improvement potential remains, but applies to reduced installed power. The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) The energy of the wind varies by region and by area. In a study that was carried out at the Technical University of Berlin [18], three different types of sail were modelled onto two types of ships on three different routes. The objective of that study was to assess the savings of energy and of fuel that might be obtainable over a period of five years, using actual weather data. This study indicated that the potential for sail energy was better in the North Atlantic and North Pacific as compared to the South Pacific. Fuel savings were slightly larger at higher speeds; however, in terms of percentages; the fuel savings were greater at low speed due to the low total demand for propulsion power. In percentage terms, savings were typically about 5% at 15 knots, rising to about 20% at 10 knots. With optimal weather routeing, these figures improved. The best ship with the best sail type, with optimal weather routeing, operating in the most favourable five-year average weather (North Atlantic), was shown to save 15% at 15 knots and 44% at 10 knots. Presently, full-scale trials are being undertaken using kites. This technology is also discussed, in the context of marginal cost and abatement potential, in appendix 4.

167 MEPC 71/INF.12 Annex 4, Page 37 Categorization / Type of (6)Solar power Short-description Naturally, it is difficult to simulate such complex systems, and currently there is limited full-scale experience with modern commercial sail ships against which such a model can be validated. Also, without such experience, it is also difficult to assess the practical feasibility of the size and number of sails modelled. The above figures should thus be considered indicative. Nevertheless, sail-assisted power does seem to be an interesting opportunity for saving fuel in the medium- and long-term picture. A2.87 When assessing the potential of solar power for application on ships, it is interesting to consider the potential available energy. Earth's average solar irradiance on the surface is approximately 342 W/m 2. On average, 30% of this radiation will be reflected back to space. Clouds are the main contributor to the reflection. The solar irradiance will vary with latitude, season, weather conditions and time of day. How much of this energy a photovoltaic cell will be able to capture depends on the efficiency of the cell and the positioning of the cell relative to the solar beam. Current solar cells have an efficiency of about 13%. Today, the best technology, which is used in laboratories and on spacecraft, has an efficiency of approximately 30%. Efficiencies are predicted to reach 45 60% when third-generation photovoltaic cells are developed and matured. The specific power of solar cells is given in table. Table Indicative specific power of solar cells Current Current best Future Approximate energy conversion efficiency (%) ratio (%) 1% reduction of power consumption for ship types with sufficient deck space available ratio when using with other energy saving (%) This could be combined with all other without reduction of the improvement ration; potential will increase in combination with (integration into) sails due to the increase of area available for panels The year expected for in-use of that technology Already available Current status of application Rarely used Applicable ships are limited because of their arrangement. This technology has been used just as trial basis. Because of the limitation of space, maintenance and cost effectiveness, this would not be applied for conventional ships. Not relevant for container vessels as only very little deck Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) Millions of dollars

168 MEPC 71/INF.12 Annex 4, Page 38 Categorization / Type of Nominal power (W/m 2 ) Power adjusted for reflection(w/m 2 ) Short-description ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application area are available. Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) To get an idea of how much power it is possible to get from photovoltaic cells on a ship, the following example calculation has been made for a tanker with a length of 270 m and a breadth of 50 m (see table A2-4). A tanker of this size is equipped with an engine that is rated to approximately 18,000 kw, and the auxiliary power would be around 1,000 kw. Table Power production by photovoltaic cells, assuming that the tanker's deck area is completely covered by Approximate energy conversion efficiency (%) Nominal power (kw) Power adjusted for reflection (kw) solar cells Current Current best Future ,406 2, ,968 Current solar-cell technology would thus, on average, only be sufficient to cover a fraction of the auxiliary power even

169 MEPC 71/INF.12 Annex 4, Page 39 Categorization / Type of Short-description if the complete deck area was covered by photovoltaic cells. Therefore, it can be concluded that, due to the limited capacity of solar cells in respect to surface area that they cover, they do not yet appear to be a very efficient source of energy supply. Furthermore, at certain times and in certain areas, solar radiation will be above average and the auxiliary power demand could be met. Also, by using highly efficient (presumably expensive) spacecraft-type solar cells, current power demand could, on average, be met. ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) Also, since solar power is not always available (e.g., at night) backup power would be needed; even if the power is available, on average, at day time, this would not help reduce the demand for auxiliary power at night unless there is an energy storage system available on board. 7 Other (1)Optimization of principle dimensions Solar energy can also be used for heating purposes, e.g. of water while the ship is in port. (Excess heat is normally available on board ships at sea.) Changing restriction of port, canal, etc. may bring the improvement of propulsion performance by changing principle dimensions. e.g. Panama canal expansion. It depends on development of infrastructure. This can be combined with some of the other without loss of improvement ratio, in particular categories 5, 6, and 7. The improvement Already available Normally used Most likely to many constrains. However, we will often look at main dimensions to find lowest transport cost. It depends on development of infrastructure. None

170 MEPC 71/INF.12 Annex 4, Page 40 Categorization / Type of (2)Reduction of light weight Short-description Increasing deadweight by reducing light weight, in keeping same displacement. ratio (%) 0-3% The effect of H- CSR should also be considered. The effect of this technology depends on ship type and ship size. ratio when using with other energy saving (%) potential of of category 1 and 7(1) might be slightly reduced. The engine / fuel related improvement potential remains, but applies to reduced installed power. This can be combined with all without loss of improvement ratio. The year expected for in-use of that technology Already available Current status of application It depends on development of infrastructure. Normally used Future prospect of the applications / adoptions of the technology Wider use of innovative lightweight materials, such as FRP, require revision and modification of IMO instruments Estimated whole cost *1 (US dollar) None (3)Hull Vane A Hull Vane has been applied on a number of superyachts. The Hull Vane is a fuel saving device in the form of a fixed foil, located below the stern of a ship. The Hull Vane influences the stern wave pattern and creates hydrodynamic lift, which is partially oriented forward. This results in a reduction in of the ship's resistance. Refer to and K. Uithof et al, "An Update on the The improvement ratio depends on the ship types and the speed. It is reported that on merchant ships, the potential This can be combined with some of the other without loss of improvement ratio, e.g. Already in use. Rarely used Hardly relevant for bigger vessels due to slow steaming.

171 MEPC 71/INF.12 Annex 4, Page 41 Categorization / Type of (4)Other energy saving applications on superyachts for Superyachts Short-description Development of the Hull Vane R ", HYPER 2014 conference. Hull Vane has also been applied for supply vessels. Application for many other merchant vessels has proven to be effective on the basis of extensive CFD analyses. 1) Energy saving lighting such as LEDs. 2) Utilizing engine waste heat where possible for items like DPF catalyzers. 3) Engine selections and guidance to operators on speed ranges to minimize BSFC. These are verified on sea trials by running measured distance runs while recording actual fuel consumed. 4) Selection and design of underwater equipment to minimize drag, including stabilization equipment, cathodic protection devices, propeller strut shapes and alignment and propeller designs. 5) Tank testing and CFD work to maximize vessel efficiencies 6) Fitting of appendages such as bulbous bows for fuel economy 7) Extensive weight estimating and weight reduction strategies to increase vessel efficiencies (including application of aluminum and composite structural materials and lightweight interior materials). 8) Optimization of power management strategies to ensure that generators are producing power at maximum efficiencies. 9) Installation of Diesel Electric equipment to match operational profiles. ratio (%) resistance reductions between 5% and 10% are common. ratio when using with other energy saving (%) categories 1, 2, 4, 5 and 6. Slightly reduced potentials can be anticipated in combination with category 2. The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar)

172 MEPC 71/INF.12 Annex 4, Page 42 Categorization / Type of Short-description <Comments: Some of the above are also applicable to other ship types, e.g. 1) for all passenger ships and can be significant; Some of the above are already covered above and are applicable for all ship types, e.g. 4), 5), 6), 7), 8) and 9)> ratio (%) ratio when using with other energy saving (%) The year expected for in-use of that technology Current status of application Future prospect of the applications / adoptions of the technology Estimated whole cost *1 (US dollar) ***

173 Annex 5, Page 1 ANNEX 5 TABLE OF CORRESPONDENCE BETWEEN ICE CLASSES OF DIFFERENT CLASSIFICATION SOCIETIES This table has developed as a reference for identifying ice class ships for which regulation 20 and 21 of MARPOL Annex VI shall not apply. This table doesn t show the comparison between ice classes of different classification societies. Classification Society Finnish-Swedish Ice Class Rules Russian Maritime Register of Shipping (Rules 2008) American Bureau of Shipping Bureau Veritas CASPPR, 1972 China Society Classification Ice Classes lower than IA Super IА IB IC II Arc 4 Ice 3 Ice 2 Ice 1 Ice Class I А Ice Class I B Ice Class I C D0 ICE CLASS IА ICE CLASS IВ ICE CLASS IC ID В С D E Ice Class B1 Ice Class B2 Ice Class B3 Ice Class B Ice Class Ice Classes equal to IA Super Ice classes higher than IA Super IА Super - Arc 5 Arc 6, Arc7, Arc8, Arc 9, Icebreaker 6, Icebreaker 7, Icebreaker 8, Icebreaker 9 Ice Class I АА А0, A1, A2, A3, A4, A5 ICE CLASS IA SUPER А - CAC4, CAC3, CAC2, CAC1 Ice Class B1* - Det Norske Veritas DNV GL Germanischer Lloyd ICE-1А ICE-1B ICE-1С ICE-C Ice(1A) Ice(1B) Ice(1C) - Е3 E2 E1 E ICE-1А* ICE-05, ICE-10, ICE-15, POLAR-10, POLAR-20, POLAR-30 Ice(1A*) - Е4 Arc1 - Arc4

174 Annex 5, Page 2 Ice Class Classification Society Ice Classes lower than IA Super Ice Classes equal to IA Super Ice classes higher than IA Super IACS Polar Rules PC7 PC6 PC5, PC4, Korean Register of Shipping Lloyd s Register of Shipping Nippon Kaiji Kyokai PolskiRejestrStatków IA IB IC ID Ice Class 1A FS (+) Ice Class 1A FS Ice Class 1B FS (+) Ice Class 1B FS Ice Class 1C FS (+) Ice Class 1C FS Ice Class 1D Ice Class 1E NS* (Class IA Ice Strengthening) NS (Class IA Ice Strengthening) NS* (Class IB Ice Strengthening) NS (Class IB Ice Strengthening) NS* (Class IC Ice Strengthening) NS (Class IC Ice Strengthening) NS* (Class ID Ice Strengthening) NS (Class ID Ice Strengthening) L1 L2 L3 L4 PC3,PC2, РС1 IA Super - Ice Class 1AS FS (+) Ice Class 1АS FS NS* (Class IA Super Ice Strengthening) NS (Class IA Super Ice Strengthening) Ice Class AC3 Ice Class AC2 Ice Class AC1.5 Ice Class AC 1 L1A - - RegistroItalianoNavale ICE CLASS IA ICE CLASS IB ICE CLASS IC ID ICE CLASS IA SUPER - ***

175 Annex 6, page 1 Analysis of 2017 EEDI Database Source: MEPC 71/INF.14 EEDI database Review of status of technological development (Regulation 21.6 of MARPOL Annex VI) Note by the Secretariat Database as of 31 March 2017 (2,443 ships) H:\MEPC\72\MEPC-72-INF-12-annex 6

176 Annex 6, page 2 General Observations There is a large amount of data for 4 ship types But the increase in number of ships compared to the 2016 database is not large Results of 4 main ship types Bulk carriers Average EEDI is only slightly better compared to 2016 database, but spread is also slightly less Only bulk carriers 40,000 to 74,900 DWT meet the 20% reduction, on average Very few bulk carriers meet the 30% reduction Tankers Average EEDI is only slightly better compared to 2016 database, but spread is also slightly less Tankers <225,000 DWT meet the 20% reduction, on average Tankers >225,000 DWT do not meet the 20% reduction Some tankers meet the 30% reduction, but most do not Containerships Average EEDI is better compared to 2016 database, but spread is larger Most containership designs meet the 20% reduction target, although there are still some designs that don t Many containerships meet the 30% reduction, but a significant number do not Gas Carriers Average EEDI is better compared to 2016 database, but spread is larger Very few gas carriers meet the 30% reduction H:\MEPC\72\MEPC-72-INF-12-annex 6

177 Annex 6, page 3 General Observations There is only a small amount of data for 4 ship types General Cargo The database is still very small: 37 ships (there are 15,776 ships in Sea-Web) Average EEDI is better compared to 2016 database, but spread is larger Refrigerated cargo: 2 ships One is 39.2% better than the reference line, the other is 12.6% better RoRo Vehicle: 23 ships, none of which are subject to reduction requirements RoRo Cargo: 9 ships, none of which are subject to reduction requirements There is no data at all for 4 ship types: Combination carrier LNG carrier RoRo Passenger Cruise ship w/nonconventional propulsion H:\MEPC\72\MEPC-72-INF-12-annex 6

178 Annex 6, page 4 Bulk Carriers Bin values are the upper bounds SeaWeb 2014: 10,755 EEDI: >10,000 H:\MEPC\72\MEPC-72-INF-12-annex DWT 6 Number of Ships Average St. Deviation 2015 data set % 6.0% 2016 data set % 5.1% 2017 data set % 4.8% only % 4.3% only % 4.3%

179 Annex 6, page 5 Bulk Carriers by sub-category Number of Ships Average Standard Deviation 2015 data set % 6.0% 2016 data set % 5.1% 2017 data set % 4.8% 10,000 39,900 DWT % 6.1% 40,000 74,900 DWT % 4.0% 75, ,900 DWT % 4.8% >175,000 DWT % 3.8% (2016 data set) % 4.3% (2017 data set) % 4.3% EEDI applies ships >10,000 DWT Only ships 40,000-74,900 DWT meet 20% on average There may be issues for the other sub-category designs H:\MEPC\72\MEPC-72-INF-12-annex 6

180 Annex 6, page 6 Tanker Bin values are the upper bounds SeaWeb 2014: 12,462 (oil and chemical) EEDI: >4,000 DWT H:\MEPC\72\MEPC-72-INF-12-annex 6 Number of Ships Average St. Deviation 2015 data set % 8.3% 2016 data set % 8.2% 2017 data set % 8.1% only % 6.9% only % 7.0%

181 Annex 6, page 7 Tankers by sub-category Number of Ships Average Standard Deviation 2015 data set % 8.3% 2016 data set % 8.2% 2017 data set % 8.1% 4,000-59,900 DWT % 7.3% 60, ,000 DWT % 7.2% >225,000 DWT % 4.8% (2016 data set) % 6.0% (2017 data set) % 7.0% EEDI applies ships >4,000 DWT Ships <225,000 DWT meet 20% improvement on average Ships >225,000 DWT don t meet it H:\MEPC\72\MEPC-72-INF-12-annex 6

182 Annex 6, page 8 Containership Bin values are the upper bounds SeaWeb 2014: 5,040 EEDI: >10,000 DWT H:\MEPC\72\MEPC-72-INF-12-annex 6 Number of Ships Average St. Deviation 2015 data set % 11.6% 2016 data set % 11.5% 2017 data set % 12.4% % 12.3% % 12.4%

183 Annex 6, page 9 Containership by sub-category Number of Ships Average Standard Deviation 2015 data set % 11.6% 2016 data set % 11.5% 2017 data set % 12.4% <10,000 DWT 2 <10,000-50,000 DWT % 11.7% 50, ,000 DWT % 13.4% 100, ,000 DWT % 10.5% >180,000 DWT % 5.6% (2016 data set) % 10.7% (2017 data set) % 12.4% EEDI applies ships >10,000 DWT Most designs meet 20% improvement or better, for the most part, but there are still some designs that don t make it H:\MEPC\72\MEPC-72-INF-12-annex 6

184 Annex 6, page 10 Gas Carrier Bin values are the upper bounds SeaWeb 2014: 1,683 EEDI: >2,000 DWT H:\MEPC\72\MEPC-72-INF-12-annex 6 Number of Ships Average St. Deviation 2015 data set % 8.2% 2016 data set % 8.8% 2017 data set % 9.5% % 8.9% % 8.9%

185 Annex 6, page 11 General Cargo Bin values are the upper bounds SeaWeb 2014: 15,776 EEDI: >3,000 DWT H:\MEPC\72\MEPC-72-INF-12-annex 6 Number of ships Average St. Deviation 2015 data set % 20.1% 2016 data set % 19.6% 2017 data set % 20.3% % 9.6% % 9.6%

186

187 Transportation Research Part D 52 (2017) MEPC 72/INF.12 Annex 7, page 1 Contents lists available at ScienceDirect Transportation Research Part D journal homepage: State-of-the-art, measures, and potential for reducing GHG emissions from shipping A review Evert A. Bouman a,, Elizabeth Lindstad b, Agathe I. Rialland b, Anders H. Strømman a a Industrial Ecology Programme and Department of Energy and Process Engineering, Norwegian University of Science and Technology (NTNU), NO-7491 Trondheim, Norway b SINTEF Ocean AS (MARINTEK), NO-7450 Trondheim, Norway article info abstract Article history: Available online 4 April 2017 Keywords: Maritime transport Shipping and the environment Greenhouse gases Abatement options Emission reductions CO 2 emissions from maritime transport represent around 3% of total annual anthropogenic greenhouse gas (GHG) emissions. These emissions are assumed to increase by % in 2050 in business-as-usual scenarios with a tripling of world trade, while achieving a C climate target requires net zero GHG emissions across all economic sectors. Consequentially, the maritime sector is facing the challenge to significantly reduce its GHG emissions as contribution to the international ambition to limit the effects of climate change. This article presents the results of a review of around 150 studies, to provide a comprehensive overview of the CO 2 emissions reduction potentials and measures published in literature. It aims to identify the most promising areas, i.e. and operational practices, and quantify the combined mitigation potential. Results show a significant variation in reported CO 2 reduction potentials across reviewed studies. In addition, no single measure is sufficient to achieve meaningful GHG reductions. Emissions can be reduced by more than 75%, based on current and by 2050, through a combination of measures if policies and regulations are focused on achieving these reductions. In terms of emissions per freight unit transported, it is possible to reduce emissions by a factor of 4 6. Ó 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( 1. Introduction For centuries, sea transport has been a major facilitator of trades between nations, regions, and continents. More recently, together with trade liberalisation, telecommunication, and international standardisation, it has been a key enabler of globalisation (Hoffmann and Kumar, 2002). Over the past 40 years, maritime transport has increased by 250%, following the same growth rate as global Gross Domestic Product (GDP), and growing more rapidly than energy consumption (170%) and global population (90%) Eskeland and Lindstad, From a global freight transport perspective, shipping is recognized as an energy-efficient means of transportation compared to road and air transport, due to its large carrying capacity and low fuel consumption per ton transported. According to the third greenhouse gas study (GHG) of the International Maritime Organization (IMO), shipping emitted 938 Mt CO 2 in 2012, accounting for 2.6% of global anthropogenic CO 2 emissions. This is a reduction compared to the 1100 Mt Corresponding author. address: evert.bouman@ntnu.no (E.A. Bouman) /Ó 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( H:\MEPC\72\MEPC 72-INF.12-annex 7

188 Annex 7, page 2 E.A. Bouman et al. / Transportation Research Part D 52 (2017) CO 2 emitted in 2007 (3.5% of global emissions) and can be attributed to the increase in vessel size and lower operational speeds (Lindstad et al., 2015a; Smith et al., 2014). Notwithstanding this amelioration, in a business-as-usual (BAU) scenario with a tripling in world trade and no further mitigating measures taken, future emissions are expected to increase by % over the period (Buhaug et al., 2009). These emission growth prospects are opposite to what is required to reach a climate target well below 2 C by the year 2100 (IPCC, 2007). Global GHG emissions have to decrease to net zero and even further to negative values across all sectors by the second half of the century. The decarbonisation level required by each of the sectors is dependent on the widespread adoption of so-called negative emissions and measures, such as bioenergy with carbon dioxide capture and storage and afforestation, to balance sources with unavoidable positive emissions. However, continuous and widespread deployment of negative emissions is currently not happening at the required scale. Consequentially, other sectors need to decarbonize on the premise that negative emissions might not work at scale (Anderson and Peters, 2016). One approach to reconcile shipping emissions with international commitments on climate change is to treat the shipping sector as if it were a sovereign nation that contributes fair and proportionate to emissions budgets. Following this approach, a reduction of at least 85% relative to 2010 is necessary by 2050 (Anderson and Bows, 2012). This implies that even in a nogrowth scenario the CO 2 emissions per unit freight transport from shipping will have to be reduced from approximately 25 to 4 g of CO 2 per ton-nautical mile, i.e. a reduction by a factor of five to six (Lindstad, 2013). Previous studies of emissions abatement measures in shipping have documented that it is possible to improve energy efficiency and reduce emissions in a cost effective manner, either with zero costs or with net cost savings (Lindstad et al., 2015a; Buhaug et al., 2009; Faber et al., 2009, 2011; Alvik et al., 2010). There are two main types of studies: (i) those estimating the total CO 2 reduction improvement potential for the global shipping sector (e.g. Buhaug et al., 2009; Lindstad, 2013; Alvik et al., 2010), and (ii) those investigating more closely one or a small group of measures together (e.g. Gilbert et al. 2014). Whereas the first type provides oversight, the second type is able to provide more detailed estimates and taking into account distinct contexts of application, such as the studies of speed and its impact on emissions (Corbett et al., 2009; Lindstad et al., 2011; Pélerin et al., 2010). Industrial Research and Development initiatives related to fuel efficiency, GHG emissions reduction, as well as policies for controlling air emissions (Lindstad, 2013; Kontovas and Psaraftis, 2016; Psaraftis, 2016; Eide et al., 2013a) have resulted in a high level of knowledge, awareness, as well as operational adaptations across the maritime sector (Dnv, 2014). However, the quantification of a (theoretical) maximum potential for energy- and GHG reducing measures is complicated by organizational, technical, economic and political barriers for implementation (Jafarzadeh and Utne, 2014; Tillig et al., 2015; Rojon and Dieperink, 2014), as well as variations in applicability of measures for distinct ship types and/or ages (Eide et al., 2013a, 2013b; Tillig et al., 2015; Franc, 2014; Lin, 2012; Guerra and Jenssen, 2014). Whereas overview studies of presenting mitigation measures and scenarios for shipping are available (e.g. Buhaug et al. (2009)), comprehensive reviews are sparse in the scientific literature. The motivation for this study is to provide a comprehensive overview of the CO 2 emissions reduction potentials and measures published in literature. We aim to identify the most promising measures, i.e. and operational practices, and quantify their mitigation potential. We are well aware of the importance of other pollutants, their impact on both health and climate, and the necessity to consider more than one type of emission when evaluating emission reduction measures (Eide et al., 2013b; Lindstad et al., 2016a, 2015b). For example, while CO 2 emissions from LNG combustion are lower than emissions from HFO combustion, the fugitive emission of methane (a potent GHG) during bunkering or due to slippage in the engine, dampens the GHG reduction potential. However, to be able to utilize the largest body of literature, which often does not consider other GHGs, we chose to focus on measures for mitigating CO 2 emissions as a proxy for GHG reduction. In this study, we identify the areas with the highest potential for reduction of emissions as a guide towards furthering the development of low-carbon shipping. In Section 2, we describe the method employed. The CO 2 reduction potentials based on studies with focus on emission reduction achievable at fleet level are presented in Section 3.1. We present the emission reduction potential identified by studies focusing on specific measures in Section 3.2. Subsequently, we discuss the results and the most promising areas regarding mitigation potential as well as uncertainties and gaps in the reviewed literature. In addition, we discuss the need for a more holistic approach, which includes the net contribution to climate change of all the exhaust gases when assessing alternative abatement options and. 2. Methodology We performed a systematic review of previous studies of fuel and CO 2 emissions reduction measures, based on a comprehensive search and analysis of published studies on shipping energy efficiency and GHG emissions. The review focused on the reported results and not on the methods used. The selection of studies was based on a qualitative assessment of the relevance of the study for estimating CO 2 emission reduction potentials from shipping by different means. We limited the scope by investigating studies published after the second GHG study of the IMO, published in 2009 (Buhaug et al., 2009), which provides the first most complete overview of emissions reduction potential and is used as main reference for many similar studies (Smith et al., 2014; Faber et al., 2011, 2010; Eide et al., 2013a; Wang et al., 2010). In order to systematically search, analyze, and categorize the results from previous studies, we broadly searched for literature containing a combination of the search terms: ship energy efficiency, GHG or CO 2 emissions from shipping, and ship performance, design, or operations. By reviewing titles and abstracts we filtered the results according to relevance with regards H:\MEPC\72\MEPC 72-INF.12-annex 7

189 410 E.A. Bouman et al. / Transportation Research Part D 52 (2017) MEPC 72/INF.12 Annex 7, page 3 to the aim of the review. We classified the publications based on: type of publication, the primary and secondary focus (e.g. fuel efficiency, CO 2 reduction, other GHGs), the scope of the potential identified (maximum achievable potential or potential from individual measure), the nature of the potential identified (quantitative or qualitative), and the type of measure (e.g. technical, operational). All available data on estimations of emission reduction potential (expressed in fuel efficiency, CO 2 or CO 2 -eq reduction) were registered, or calculated based on evidence in the study. In addition, we identified articles providing more insight on emissions and energy efficiency in shipping, though without quantitative estimates of reduction potential. The literature review converged on approximately 150 studies. Sixty (60) of these studies provide quantitative estimates of CO 2 emission reduction potential, of which fourteen (14) cover the maximum abatement potential at a global or regional fleet level as shown in Table 1, and the rest focuses on one or few individual measures. The remainder of studies provide qualitative complementary insights. 3. Results 3.1. Maximum potential reduction of CO 2 emissions from shipping This section presents the results from studies focusing on emission reductions achievable at fleet level, based on adoption of available measures within a given time perspective. The main study of potential emission reduction from shipping is the Second IMO GHG Study from 2009 (Buhaug et al., 2009), used as reference for follow-up studies either in terms of potential estimates for individual reduction measures or fleet development scenarios. Efforts post-2009 to assess the global and large scale reduction potential for GHG emissions are led by several SNAME and IMAREST investigations, as well as studies initiated by DNV (now DNV GL) Alvik et al., 2010; Eide et al., 2013a., 2013b; Eide et al., 2011; Hoffmann et al., In addition, two studies financed by the European Commission s Directorate General Climate Action provide valuable input to the stateof-the-art and insight on potential for reduction of emission from European shipping (Lindstad et al., 2015a; Kollamthodi et al., 2013). Based on reviews of mitigating measures (Lindstad et al., 2015a; Buhaug et al., 2009; Alvik et al., 2010; Faber et al., 2011; Tillig et al., 2015; EMEC, 2010; CNSS, 2011), as well as ongoing industrial projects and in-house knowledge, six groups of measures with high mitigation potential were selected and used to sort and compare the selected studies. These six main groups are: hull design; economy of scale; power and propulsion (including energy saving devices); speed; fuels and alternative energy sources; weather routing and scheduling. Hull design covers aspects related to hull dimensions, shape and weight, which contribute to improve the hydrodynamic performance and minimize resistance. Economies of scale is another means of reducing emissions, since larger ships and cargoes tend to be more energy-efficient per freight unit. Typically, when cargo-carrying capacity is doubled, the required power and fuel consumption increases by about two-thirds, thus reducing fuel consumption per freight unit. Power and propulsion includes design of power system and machinery, hybrid power solutions, higher propulsion efficiency, waste heat recovery, and reduction of on-board power demand by energy saving devices such as kites and sails. Hybrid power systems enable the efficient exploitation of various energy sources, such as combining batteries with combustion engines to utilize the best of each technology, i.e. batteries can be used as a buffer for covering peak power requirements and to avoid low power operations of combustion engines. Speed relates to the operational speed of the vessel, as well as its design speed. Traditionally, ships are often designed to operate at their hydrodynamic boundary speeds, i.e. the speed at which for a given hull the resistance curve starts to rise rapidly with increasing speeds. As power requirement is proportional to the product of speed and resistance, this implies that when a ship reduces its speed, the fuel consumption is reduced and that the biggest fuel reductions are achieved when ships reduce speed in the boundary area. Fuels and alternative energy sources covers all aspects related to substituting or complementing bunker fuels HFO-MGO with alternative energy carriers. Emissions of CO 2 can be cut by switching to fuels with lower total emissions, both directly and throughout the fuel cycle including production, refining and distribution. Examples are LNG and biofuels. Hydrogen is gaining increasing attention, as well as renewable energy such as wind and solar. Weather routing and scheduling consists of finding the optimum sailing route and speeds, taking into account current, wave and weather conditions, and deliveries according to the contractual agreements or published schedules, to minimize resistance and fuel consumption. Table 1 shows the potential CO 2 emission reduction at fleet level relative to the estimated BAU emissions for each individual study. The first column gives the reference to the study; the second the applied method; the third the coverage; the following six columns are used for the investigated measures where x means that the measure has been included in the study. Then from column 10 follows: fuel price used; fleet reference year; final year of scenario; BAU emissions in the final year; Identified CO 2 emission reduction potential at fleet level in the final year relative to BAU emissions in the final year. In addition, Table 1 includes the median values across all studies for BAU emission estimates as well as minimum and maximum estimated CO 2 reduction potential. The main observations from Table 1 are: the maximum potential for emission reduction spans from 20% to 77%, not including nuclear power as an alternative power source. When including nuclear power as alternative power source, maximum reduction potential reaches 95% (Eide et al., 2013a). Median reduction rates are 35%, 39% and 73% for the years 2020, 2030, and 2050 respectively. The highest range of reduction potential is found in the 2050-estimates, 58 77% (Buhaug et al., 2009; Lindstad, 2013; Eide et al., 2013a), reflecting the time it takes to adopt emission reduction measures. Apart from the H:\MEPC\72\MEPC 72-INF.12-annex 7

190 Table 1 Potential reduction in CO 2 emissions from shipping at fleet level, all measures combined. MEPC 72/INF.12 Annex 7, page 4 Author(s) Buhaug et al. (2009) Method for calculation of potential Expert estimates, fleet development scenario model 25 measures Faber et al. (2009) Same as Buhaug et al. (2009) 29 measures Alvik et al. (2010) Wang et al. (2010) Expert estimates 25 measures Similar to Buhaug et al. (2009) 22 measures Application 18 ship types, world fleet 9 ship types, 40,055 ships 59 ship segments, world fleet 14 ship types, world fleet Hull design x x Economy of Scale x Power& Propulsion x x Speed x x Fuels & alt. Energy Sources x Routing & Scheduling x x Fuel price ($/ton) Fleet reference year Final year of scenario BAU emissions in final year (Mt CO 2 ) Identified CO 2 emission reduction potential (%) 16 34% 58 75% x x x x x % x x x x x % x x x x x Faber et al. (2011) Updated of Wang et al. (2010) 14 ship types x x x x x 700, 900 Eide et al. (2011) Hoffmann et al. (2012) CCNR (2012) CO 2 emission scenario model 25 measures Cost-effectiveness. 25 measures Model-based. 27 measures Eide et al. (2013a) Same as Eide et al. (2011) 27 measures Eide et al. (2013b) Same as Eide et al. (2011) 15 measures (existing ships) Lindstad (2013) Kollamthodi et al. (2013) Heitmann and Peterson (2014) TNO Lindstad et al. (2015a) Model-based, 5 main measures Scenario model, 3 main measures Abatement costs scenarios, 22 measures Model-based, 12 measures (existing ships) 59 ship types, world fleet, 59,800 ships 59 ship segments, world fleet Inland navigation, Western Europe 59 ship segments, world fleet, 59,800 ships 59 ship segments, world fleet, 59,800 ships 11 ship types, 109,000 bt nm x x x x x % 35 45% 20 46% 20 46% x x x x x % x x x x x x % x x x x x % 56% % x x x x x % x x x x European shipping x x ship types x x x x x ship types, world fleet x x x x x x Median H:\MEPC\72\MEPC 72-INF.12-annex % 76% 20% 33% % % % 39% 73% E.A. Bouman et al. / Transportation Research Part D 52 (2017)

191 412 E.A. Bouman et al. / Transportation Research Part D 52 (2017) MEPC 72/INF.12 Annex 7, page 5 nuclear alternative, the 2050 figures are in line with the estimated maximum potential from the Second IMO study (Buhaug et al., 2009). Several factors seem to affect the estimated potential for reduction in different ways. First the number and types of measures included in the studies are directly affecting the maximum potential reduction. In addition, it is important to keep in mind that not all reduction measures are additive and can be applied simultaneously. Second, the categorization and aggregation of measures is challenging to compare across the studies. Even across studies including the same set of measures we observe variations in estimated reduction potentials. Third, the baseline year and baseline emissions (2020, 2030, and 2050) affect the magnitude of potential emission reduction. The further ahead in time, the larger the estimated potential for improvement, with a range of 33 48%, 30 56%, and 58 95% respectively for 2020, 2030, and The fuel cost given in the fifth last column has no direct effect on the total maximum (technical) potential for emission reduction, but rather on the calculation of cost-efficient potential and no-regret potential. In addition, fuel price is often a determinant for fuel choices in more economically oriented future scenarios (Lindstad et al., 2015a; Kesicki and Strachan, 2011). Table 1 shows a large variation in fuel prices, especially for the more recent studies (Lindstad et al., 2015a; Eide et al., 2013a), highlighting the importance of assessing fluctuations in energy price in order to determine its impact on investment decisions for mitigating measures and varying the threshold for cost efficient measures. The annual fleet-wide CO 2 emission estimates available in the studies presented in Table 1 are presented in Fig. 1. For each scenario reported in the reviewed studies, the BAU emissions are plotted as well as the low and high emission reduction estimates. To enhance clarity, we shaded the area between low and high emissions estimates for both BAU and reduction scenarios, and added jitter along the x-axis to distinguish between overlapping data points. It is important to stress here that reduction estimates are not solely based on adoption of technical and operational measures, but also reflect the assumptions of the reviewed articles with respect to growth in demand for transport. For example, Buhaug et al. use projections of transport in ton-nautical mile based on the IPCC scenario family (Buhaug et al., 2009), whereas Eide et al. use a fleet development model to estimate size of the fleet based on estimations of growth and scrapping rates (Eide et al., 2013a, 2011). Fig. 1 shows an overlap between BAU and reduction scenarios in 2020 and in 2030, reflecting uncertainty in the scenarios as well as different assumptions on the rate of adoption of emission mitigation measures and growth rates of global maritime transport. In the long term, there are no overlaps between BAU and reduction scenarios. However, compared to the 2012 situation (938 Mt CO 2 ), some of the reduction scenarios amount to a net increase in emissions, whereas other reduction scenarios report a net decrease in annual CO 2 emissions. This is especially clear for the period up to 2030 where one reduction scenario estimates an increase in emissions of more than 65%, where others estimate a decrease of maximum 34%. This signifies the challenge to reduce the absolute emissions of the maritime sector while transport volumes increase. Each increase in transport volume will have to be matched by larger improvements in environmental performance to achieve absolute emissions reduction Estimated CO 2 reduction potential from individual measures This section presents the estimated reduction potential from individual measures. Maritime emission and reduction measures are commonly divided into two main categories: technical and operational (Psaraftis, 2016). Technical measures focus for example on energy savings through improved energy efficient design, improved propulsion and power system, and alternative or cleaner fuels. Some measures can be applied as retrofit measures, while others can only be considered for new ships. Operational measures aim at reducing emissions during operations at ship or fleet level. Examples are optimizing speed, voyage planning, fleet management, and on-board energy management. Operational measures are adequate for any ship type, existing or new-built. We have excluded market-based measures from the present review as ultimately they Fig. 1. Annual CO 2 emissions from the global shipping fleet, distinguished by business-as-usual and reduction scenario pathways. Jitter is added to distinguish different data points. H:\MEPC\72\MEPC 72-INF.12-annex 7

192 Annex 7, page 6 E.A. Bouman et al. / Transportation Research Part D 52 (2017) provide an incentive to implement technical and operational measures. Reviews and investigations of such measures can be found in references (Eskeland and Lindstad, 2016; Psaraftis, 2016, 2012; Miola et al., 2011; Russell et al., 2010; Davidson and Faber, 2012; Lindstad et al., 2015c; Carr and Corbett, 2015; Balland et al., 2015; Cullinane and Bergqvist, 2014; Nikolakaki, 2013; Lee et al., 2013). The main findings on reductions potential at individual measure level are summarized in Table 2. The table presents twenty-two (22) measures or type of measures, for which sufficient, reliable and comparable data are available in the selected literature. These measures are grouped under five main categories: hull design, power and propulsion, alternative fuels, alternative energy sources, and operations. Compared to Table 1, this implies that Economies of scale now is one of the measures grouped under hull design; that alternative fuels and alternative energy each are main categories; that speed and weather routing and scheduling are grouped under operations. All references with quantitative estimates of emission reduction potential are given in Table 2. In addition, references providing complementary qualitative insight into the mitigation potential of individual measures are listed in the last column. The range of potential CO 2 emission reduction in percentage represents the absolute minimum and maximum value collected from the literature. Complementary to Table 2, Fig. 2 shows the CO 2 reduction potential for each of the 22 measures. A solid bar indicates the typical reduction potential area, i.e. from 1 st to 3 rd quartile of the dataset, and a thin line indicates the whole spread, which and corresponds to the ranges presented in Table 2. In addition, the data points found in each of the studies are plotted using a small circle. Fig. 2 thus gives a graphical overview of the distribution of reduction potentials presented in the literature. From Table 2 and Fig. 2 we observe a large ranges in emission reduction potential per measure reported by the individual studies. Some of the variability can be explained by differences in assumptions and benchmarks across the selected studies, but it also indicates large uncertainty as to the effectivity of reported reduction potentials. For many measures, the highest reported reduction potential is several times larger than the median value. Eight (8) out of the 22 measures have a 3 rd quartile reduction potential of 20% or higher, eight (8) out of the 22 measures have a 3 rd quartile reduction potential between 10% and 20 % and six (6) out of the 22 measures have a 3 rd quartile reduction potential of less than 10%. Hull design measures focus primarily on utilizing economies of scale and reducing resistance during operation. The results indicate that novel hull design can contribute considerably to CO 2 emissions reduction. Increasing vessel size reduces emissions per unit transport work and optimizing hull shape for reduced drag can significantly reduce power consumption and consequentially emissions. Additional measures, such as light-weighting, hull coating and lubrication can contribute to improving the performance of hulls further, but their potential as a single measure is limited. For reduction measures within the power and propulsion system, Fig. 2 shows that some studies report considerable emission reductions. However, the median estimated reduction for measures in power and propulsion system is relatively low. This reflects the challenges and boundary conditions related to implementation of these measures. For example, a truly integrated hybrid drivetrain deviates significantly from a conventional set-up. Meeting all conditions required for optimal implementation, and high emissions reductions, is challenging for such an early stage technology. For some of the more conventional measures, such as efficiency increasing devices, the past decades have already seen considerable improvements and it is likely that further improvements remain marginal as the physical limits are approached. Fig. 2 shows the highest CO 2 emissions reduction potential for the use of biofuels. However, the systemic effects of large scale adoption of biofuels reach well beyond reduction in CO 2 emissions during combustion. There are two main factors influencing the CO 2 reduction potentials of biofuels. First, the bio feedstock differs in type and quality and is processed in different manners. Variations in CO 2 reduction potential occur due to changes in feedstock, processes, efficiencies, etc. The second factor pertains to the way reduction potential is calculated. Traditionally, emissions from biological origin are assumed to be carbon-neutral as biofuel is of renewable origin and carbon is sequestered during growth of the biomass. However, the carbon-neutrality assumption depends strongly on the rotation periods of the source crop, location of the crops, and direct and indirect albedo changes due to harvesting, all of which have a climate effect. In addition, nonclimatic concerns such as competition for scarce land resources make a comparison in terms of only CO 2 emissions overly simplified (Cherubini et al., 2013). Switching to LNG as a fuel can lead to relatively high emissions. Consisting mostly of methane, the CO 2 emissions during LNG combustion are considerably lower than those of other fossil fuels and this is reflected in the reduction potential. However, leakage of methane from the engine, a potent GHG, could pose a challenge not captured in Fig. 2. In addition, as a carbon based fuel of fossil origin, combustion of LNG still results in continued CO 2 emissions. Considering that the residence time of CO 2 in the atmosphere is thousands of years (Archer et al., 2009), and that there is a clear carbon budget associated with the goals set forward in the Paris agreement, a one-sided focus on LNG as a mitigation option risks lock-in of the sector into a high-carbon infrastructure not commensurate with required commitments in the long term (Gilbert, 2014). For the measures focusing on alternative sources of energy we observe high reduction potentials for wind power and low potential for solar power. The utilization of sails, kites, and photovoltaic cells to capture these additional energy sources is strongly dependent on the ship case in which the technology is applied. Such measures are most efficient for smaller ship sizes on specific routes with high solar incidence and wind potential, as the total amount of energy that can be generated by these measures on-board is constrained by the surface area necessary for each of these measures. Conversely, cold ironing can theoretically be applied to ships of any size. It can reduce local air pollution considerably, especially in countries with clean electricity mixes. However, there appears to be little agreement between studies as to its CO 2 reduction potential, which is necessarily a function of the fraction of travel time spent in port. Few data are available in the review for application H:\MEPC\72\MEPC 72-INF.12-annex 7

193 Table 2 Measures and potential effect on energy efficiency and emissions reduction (CO 2 ). Type of measure Main measures reviewed Short description Hull design Vessel size Economy of scale, improved capacity utilization Power & propulsion system Hull shape Lightweight materials Air lubrication Resistance reduction devices Ballast water reduction Hull coating Hybrid power/ propulsion Power system/machinery Propulsion efficiency devices Waste heat recovery On board power demand Dimensions & form optimization High strength steel, composite Hull air cavity lubrication Other devices/retrofit to reduce resistance Change in design to reduce size of ballast Distinct types of coating Hybrid electric auxiliary power and propulsion (Incl. e.g. variable speed electric power generation) On board or auxiliary power demand (e.g. lighting) Potential CO 2 reduction References to studies providing estimates 4 83 % Lindstad (2013), Faber et al. (2011), Gilbert et al. (2014), Tillig et al. (2015), Wang et al. (2010), Miola et al. (2011), Wärtsila (2009), Stott and Wright (2011), Lindstad et al. (2012, 2016b), Gucwa and Schäfer (2013), Lindstad and Eskeland (2015), Halfdanarson and Snåre (2015), Pauli (2016), Zöllner (2009) 2 30% Lindstad et al. (2015a, 2013a, 2014), Buhaug et al. (2009), Lindstad (2013), Faber et al. (2011), Gilbert et al. (2014), Tillig et al. (2015), Lin (2012), Wang et al. (2010), CCNR (2012), Miola et al. (2011), Wärtsila (2009), Stott and Wright (2011), Lindstad and Eskeland (2015), Ulstein (2009) % Buhaug et al. (2009), Faber et al. (2011), Tillig et al. (2015), Wang et al. (2010), CCNR (2012), Miola et al. (2011), Wärtsila (2009), Hertzberg (2009) 1 15% Buhaug et al. (2009), Faber et al. (2009, 2011), Tillig et al. (2015), Wang et al. (2010), CCNR (2012), Miola et al. (2011), Wärtsila (2009), Wang and Lutsey (2013) 2 15% EMEC (2010), CCNR (2012), Miola et al. (2011), Wärtsila (2009) 0 10% Lindstad et al. (2015a), Tillig et al. (2015), Miola et al. (2011), Wärtsila (2009) 1 10% Buhaug et al. (2009), Faber et al. (2009, 2011), Lin (2012), Wang et al. (2010), Miola et al. (2011), Wärtsila (2009), Wang and Lutsey (2013), Maddox Consulting (2012) 2 45% Lindstad et al. (2015a), Faber et al. (2011), Tillig et al. (2015), Wang et al. (2010), CCNR (2012), Wärtsila (2009), Lindstad and Sandaas (2014, 2016), Sciberras et al. (2015) 1 35% Buhaug et al. (2009), Faber et al. (2011), Tillig et al. (2015), Lin (2012), Wang et al. (2010), CCNR (2012), Miola et al. (2011), Wärtsila (2009), Wang and Lutsey (2013), Maddox Consulting (2012), Baldi (2013) 1 25% Lindstad et al. (2015a), Buhaug et al. (2009), Faber et al. (2009, 2011), Gilbert et al. (2014), Psaraftis (2016), Tillig et al. (2015), Lin (2012), Wang et al. (2010), CCNR (2012), Wärtsila (2009), Wang and Lutsey (2013), Maddox Consulting (2012) 1 20% Lindstad et al. (2015a), Faber et al. (2009, 2011), Gilbert et al. (2014), Psaraftis (2016), Tillig et al. (2015), Lin (2012), Wang et al. (2010), EMEC (2010), CCNR (2012), Wärtsila (2009), Wang and Lutsey (2013), Maddox Consulting (2012), Baldi (2013), Future (2012), Baldi et al. (2013), Choi and Kim (2013a, 2013b), Baldi and Gabrielii (2015), Deniz (2015) 0.1 3% Lindstad et al. (2015a), Faber et al. (2009, 2011), Tillig et al. (2015), Wang et al. (2010), Wärtsila (2009), Wang and Lutsey (2013), Maddox Consulting (2012) Additional studies MEPC 72/INF.12 Annex 7, page 7 Buhaug et al. (2009), Lindstad (2013, 2015), Cullinane and Khanna (2000), Sys et al. (2008), Wu and Lin (2015), Styhre (2010), Bittner et al. (2012) Job (2015), Sánchez-Heres (2015), Shipping and Marine (2015) Doulgeris et al. (2012), Solem et al. (2015) 414 E.A. Bouman et al. / Transportation Research Part D 52 (2017) H:\MEPC\72\MEPC 72-INF.12-annex 7

194 Alternative fuels Biofuels 25 84% Lindstad et al. (2015a), Faber et al. (2009, 2011), Gilbert et al. (2014), Eide et al. (2013a), Wang et al. (2010), Bengtsson et al. (2012), Brynolf et al. (2014a) LNG 5 30% Lindstad et al. (2015a), Buhaug et al. (2009), Faber et al. (2009, 2011), Gilbert et al. (2014), Psaraftis (2016), Eide et al. (2013a), Wang et al. (2010), CNSS (2011), Baldi et al. (2013), Bengtsson et al. (2012), Brynolf et al. (2014a, 2014b), Einang (2009), Seddiek (2015) Alternative energy sources Operation Wind power Kite, sails/wings 1 50% Lindstad et al. (2015a), Buhaug et al. (2009), Faber et al. (2009, 2011), Gilbert et al. (2014), Psaraftis (2016), Tillig et al. (2015), Wang et al. (2010), EMEC (2010), CNSS (2011), Wärtsila (2009), Wang and Lutsey (2013), Clauss et al. (2007), Smith et al. (2013), Traut et al. (2014), Teeter and Cleary (2014) Fuel cells 2 20% Lindstad et al. (2015a), Faber et al. (2009, 2011), Gilbert et al. (2014), Kotb et al. (2013) Cold ironing Electricity from shore 3 10% Lindstad et al. (2015a), Gilbert et al. (2014), Lin (2012), CCNR (2012), Miola et al. (2011), Chatzinikolaou and Ventikos (2013) Solar power Solar panels on deck % Lindstad et al. (2015a), Buhaug et al. (2009), Faber et al. (2009, 2011), Gilbert et al. (2014), Wang et al. (2010), CNSS (2011), Wärtsila (2009), Wang and Lutsey (2013), Sjöbom and Magnus (2014) Speed optimization Capacity utilization Voyage optimization Other operational measures H:\MEPC\72\MEPC 72-INF.12-annex 7 Operational speed, reduced speed At vessel and fleet level (fleet management) Advanced weather routing, route planning and voyage execution Trim/draft optimization, Energy management, Optimized maintenance 1 60% Lindstad et al. (2015a, 2011, 2016b, 2013b), Buhaug et al. (2009), Lindstad (2013), Faber et al. (2009, 2011, 2010), Gilbert et al. (2014), Corbett et al. (2009), Tillig et al. (2015), Lin (2012), Wang et al. (2010), CNSS (2011), CCNR (2012), Miola et al. (2011), Wärtsila (2009), Lindstad and Eskeland (2015), Wang and Lutsey (2013), Maddox Consulting (2012), Chatzinikolaou and Ventikos (2013), Norlund and Gribkovskaia (2013) 5 50% Buhaug et al. (2009), Lindstad (2013), Faber et al. (2009), Gucwa and Schäfer (2013), Lindstad et al. (2016b) % Buhaug et al. (2009), Faber et al. (2009, 2011), Wang et al. (2010), Miola et al. (2011), Wärtsila (2009), McCord et al. (1999) Lindstad et al. (2015a, 2013b), Lindstad (2013), Psaraftis (2016), Tillig et al. (2015), Lin (2012), CCNR (2012), Wang and Lutsey (2013), Maddox Consulting (2012), Johnson and Styhre (2015) 1 10% Buhaug et al. (2009), Faber et al. (2009, 2011), Tillig et al. (2015), Lin (2012), Wang et al. (2010), CCNR (2012), Miola et al. (2011), Wärtsila (2009), Wang and Lutsey (2013), Maddox Consulting (2012), Seddiek (2015) MEPC 72/INF.12 Annex 7, page 8 Brynolf et al. (2014a, 2014b), Bengtsson (2011), Bengtsson et al. (2014), Kristensen (2012), Grahn et al. (2013), Taljegard et al. (2014) Jafarzadeh et al. (2012), AEsoy et al. (2011), Einang (2007), Chryssakis et al. (2014), Thomson et al. (2015) Nuttall (2013), B9Shipping (2016), Ecoliner (2016), SkySails (2015), Schmitz and Madlener (2015), Dadd et al. (2011) Alvik et al. (2010), Welaya et al. (2013), Ludvigsen and Ovrum (2012) Alvik et al. (2010) Qiu et al. (2015), Nuttall (2013), Cotorcea et al. (2014), Lock (2013), Glykas et al. (2010) Psaraftis and Kontovas (2013), Yin et al. (2014), Woo and Moon (2014) Fagerholt et al. (2009, 2010), Bausch et al. (1998), Álvarez (2009), Fagerholt (2001), Norstad et al. (2011), Kontovas (2014) Johnson and Styhre (2015), Ranheim and Hallet (2010), Rialland et al. (2014), Poulsen and Sornn-Friese (2015) E.A. Bouman et al. / Transportation Research Part D 52 (2017)

195 416 E.A. Bouman et al. / Transportation Research Part D 52 (2017) MEPC 72/INF.12 Annex 7, page 9 Fig. 2. CO 2 emission reduction potential from individual measures, classified in 5 main categories of measures. of fuel cells for power generation. As an additional means of power generation, its effects might be marginal, but the question remains if a system could be designed in which most power comes from a fuel cell. As ship resistance is directly correlated with shipping speed, it is no surprise that speed optimization is another measure where relative high reductions in fuel consumption and emissions can be achieved. However, many data points lie well beyond the 3 rd quartile boundary, suggesting low agreement in the literature on the reduction potential. Capacity utilization shows a high median value, but the large range as well as relatively low amount of data points increases the uncertainty towards its potential as a reduction measure. For voyage optimization, most studies indicate a potential less than 10 percent, but few outliers report higher potentials well beyond the 3 rd quartile boundary. H:\MEPC\72\MEPC 72-INF.12-annex 7

196 Annex 7, page 10 E.A. Bouman et al. / Transportation Research Part D 52 (2017) If all options depicted in Fig. 2 could have been combined, which is a highly theoretical exercise, the emission reductions would be over 99% based on 3 rd quartile values, 96% based on the median, and 80% based on 1 st quartile values. However, many of the presented options are mutually exclusive. In addition, some of the reported reduction potentials are not directly additive due to interdependency of the measures. Nonetheless, there is a large number of practical and economically feasible combinations of measures. One of these combinations would be: Vessel size; Hull shape; Ballast water reduction; Hull coating; Hybrid power/propulsion; Propulsion efficiency devices; Speed optimization and Weather routing. Assuming relatively large independence between the individual measures, combining these options can lead to emission reductions of 78% based on 3 rd quartile values, 55% based on the median, and 29% based on 1 st quartile values. Introducing alternative fuels to the above combination of measures would increase the reduction potential to 67 88% and 85 96% for respectively median and 3 rd quartile values, and depending on the mix between LNG and biofuels. However, the reduction potential in the use of biofuels is based on the assumption of carbon neutrality of fuel from biological origin. In total these results indicate that there is a potential for reducing emissions by %, i.e. a factor of 4 6 per freight unit transported, based on current and based on the 3 rd quartile values. If we instead use the median values, an emission reduction of between 50 and 60% per freight unit transported up to 2050 is more realistic. 4. Discussion This study has reviewed over 150 studies to provide a comprehensive overview of the GHG emissions reduction potentials for maritime transport and measures published in literature. Its aim was to identify the most promising areas, i.e. and operational practices and quantify their combined mitigation potential. A direct comparison of the results presented in the reviewed studies comes with inherent uncertainty, as methodologies across studies differ and reduction potentials are identified with respect to different baselines. The CO 2 reduction potentials based on studies with focus on emission reduction achievable at fleet level in Section 3.1 range from 33 to 77% against a 2050 scenario baseline, indicating a high potential. Moreover, if the emission reduction is based on studies focusing on specific measures as in section 3.2, this potential increases to 80% or more based on the studies that identified the highest emission reduction potentials, i.e. 3 rd quartile values. If the 3 rd quartile values are replaced with the median, the reduction potential is reduced to 50% and upward per freight unit transported. These numbers are in line with a previous study by Gilbert et al. investigating the reduction potential of combined measures to decarbonize vessels. Newly built bulk carriers and tankers are quantified with a potential in the range of 81 98%, and containers ships with an even higher range of 92 98%. It should be noted, however, that upper bounds assume high reductions through use of biofuels (Gilbert et al., 2014). Three main conclusions can be drawn: first, it is possible to reduce emissions per freight transport unit by 75% and above up to Second, reaching such a level is based on the studies that showed reduction potential in the high end, i.e. the 3 rd quartile values. Third, achieving such high reduction is necessary to ensure absolute reductions in annual CO 2 emissions of the sector, as the continued future growth of maritime transport offsets the gains made in individual cases. In addition, a remaining challenge is to be able to realize the required GHG reductions, while at the same time meet customer demands and remain competitive in comparison to other transport modes, i.e. road, rail, and aviation. CO 2 is not the only greenhouse gas emitted by the shipping sector. Surprisingly few studies focusing on GHGs cover more than carbon dioxide (Smith et al., 2014; Eide et al., 2013b; Sciberras et al., 2015; Baldi et al., 2013; Brynolf et al., 2014b; Seddiek, 2015; Paxian et al., 2010; Lack and Corbett, 2012; Ma et al., 2012). Studies considering tighter regulations of NO x and SO x emissions in the ECAs rarely focus on the impact on CO 2 emissions or the total GHG effect. It appears that, for most key pollutants, emissions reduction go hand in hand with reductions in CO 2 emissions, e.g. through a combination of fuel switching (i.e. to low sulphur fuel or natural gas) and abatement equipment. However, several authors report an increase in CO 2 equivalent emissions as a function of stricter NO x and SO x regulations (Eide et al., 2013b; Lindstad et al., 2016a, 2015b; Gilbert, 2014; Ma et al., 2012). Current regulations provide emission limits for CO 2 for its climate change effects and for NO x and SO x for their health and environmental effects (Eide et al., 2013b). This represents a conflict, since the NO x and SO x emissions that are regulated for environmental reasons tend to mitigate global warming (Lauer et al., 2007; Eyring et al., 2010), while the unregulated emissions, i.e., BC and methane, contribute to global warming (Lindstad and Sandaas, 2016; Jacobson, 2010; Bond et al., 2013; Myhre et al., 2013; Fuglestvedt et al., 2014). These effects are not captured by the current study, but do warrant further investigation. The ranges and grouping of data points presented in Fig. 2 give an indication on the level of agreement between studies, while the total number of data points indicate the amount of evidence. Together, agreement and amount of evidence can be used to express and communicate confidence levels in a qualitative way, where the highest level of confidence is given for measures with high agreement and robust evidence (Mastrandrea et al., 2010). While reviewing a sizeable total number of studies, there are fewer articles per measure. Combined with relatively low levels of agreement between studies, this affects negatively the confidence with which high-potential measures can be selected. Based on the above, there is a need for more transparent research that is able to address simultaneously the climate mitigation potential and other environmental issues related to international maritime transport. For example, the climate effects of a shift from HFO to LNG including the interplay between reduced SO 2 emissions, reduced CO 2 emissions, and fugitive methane emissions are not well understood. The uncertainty and variation in the mitigation options presented in the H:\MEPC\72\MEPC 72-INF.12-annex 7

197 418 E.A. Bouman et al. / Transportation Research Part D 52 (2017) MEPC 72/INF.12 Annex 7, page 11 previous sections have to be reduced in order to aid the maritime community in making the best choices for sustainable development of maritime transport. Research has to address multiple aspects simultaneously in order to prevent double counting of efficiencies and measures. This allows for a closer look at both individual measures and their feasibility of application, as well as a combination of measures in hybrid solutions. In addition, expanding the scope to a wider transport network would provide a better picture of the environmental impact and potential for reduction of emissions from maritime transportation in relation to other transport modes and with regard to total impact from transport. Such a multimodal and logistics perspective has been raised by Lindstad (2013), Lindstad et al. (2016b), Bergqvist and Cullinane (2013), Cui and Li (2015), Cullinane (2014). While the shipping sector has engaged in establishing policies to reduce emissions, the rate of implementation and level of commitment implies that it is likely that more policies and regulations are needed to achieve the high emission reductions (Lindstad, 2013; Gilbert et al., 2014). The Energy Efficiency Design Index (EEDI) from the International Maritime Organization (IMO) is one example of such legislation. The EEDI, which now is applicable for all vessels built from 2013 onwards, is an energy efficiency requirement that puts thresholds on the CO 2 emitted per ton of goods transported for a fully loaded vessel as a function of its size and its type. These threshold values will gradually become stricter and the only way to meet the requirement will be to improve the design, the power system, or through adopting low-carbon fuels. A push for tougher baselines will hence contribute to larger emission reductions. 5. Conclusion To limit the rate of global warming the maritime sector is faced with the challenge to drastically reduce its GHG emissions in the coming decades. Sector-wide reductions are further complicated by the projected increased demand for maritime transport services. To reduce emissions, the sector has many technological and operational measures at its disposal. In this article, we extensively reviewed the literature to identify measures with high CO 2 reduction potential and quantify the range of reported reduction potentials for 22 individual measures, as well as the maximum reduction potential identified by literature describing fleet-wide reduction scenarios. Our results indicate that no single measure is sufficient by itself to reach considerable sector-wide reductions. Though there are single measures for which high reduction potentials are reported (e.g. the use of biofuels or speed optimization), the wide range of identified potentials suggests only moderate agreement across studies. To present a more balanced analysis we focused on median and 3 rd quartile values within the reviewed dataset. We identified eight measures with a 3 rd quartile potential of 20% or higher, eight measures with a 3 rd quartile potential between 10% and 20%, and six measures with a 3 rd quartile potential of less than 10%. Based on the reviewed studies, we conclude that a significant emission reduction over 75% is achievable by swift adoption and combination of a large number of individual dependent and independent measures. In other words, it is possible to reduce GHG emissions by a factor of 4 6 per freight unit transported with current within As an example, we presented one combination of measures leading to emission reductions of 78% based on 3 rd quartile values. However, many more combinations are possible and we hope that this article can act as a starting point for identifying successful combinations of emissions reduction measures. The overall success of these emissions reductions and measures is dependent on the growth rates of maritime transport. Policies, regulations, and legislation, such as the EEDI, can facilitate reduction of GHG emissions by the sector, but successful implementation has to be supported by high-quality studies addressing multiple effects and measures simultaneously in order to avoid counteracting and inefficient adoption of mitigation measures. Acknowledgements The authors gratefully acknowledge financial support from the Research Council of Norway through the Centre for Research based Innovation (SFI) Smart Maritime - Norwegian Centre for improved energy-efficiency and reduced emissions from the maritime sector. Project nr /O30. We would like to thank the anonymous reviewers for their comments and suggestions towards improving this article. References AEsoy, V., et al., LNG-Fuelled Engines and Fuel Systems for Medium-Speed Engines in Maritime Applications. SAE Technical Paper. JSAE (SAE ). Álvarez, J.F., Joint routing and deployment of a fleet of container vessels. Maritime Econ. Logist. 11 (2), Alvik, S., et al., Pathways to Low Carbon Shipping-Abatement Potential Towards 2030, DNV, Editor. Høvik. Anderson, K., Bows, A., Executing a Scharnow turn: reconciling shipping emissions with international commitments on climate change. Carbon Manage. 3 (6), Anderson, K., Peters, G., The trouble with negative emissions. Science 354 (6309), Archer, D. et al, Atmospheric lifetime of fossil fuel carbon dioxide. Ann. Rev. Earth Planet. Sci. 37, B9Shipping, < [Cited 2016 March, 1st 2016] Baldi, F., Improving ship energy efficiency through a systems perspective. In: Department of Shipping and Marine Technology. Chalmers University of Technology, p. 78. Baldi, F., Gabrielii, C., A feasibility analysis of waste heat recovery systems for marine applications. Energy 80, H:\MEPC\72\MEPC 72-INF.12-annex 7

198 Annex 7, page 12 E.A. Bouman et al. / Transportation Research Part D 52 (2017) Baldi, F., Bengtsson, S., Andersson, K., The influence of propulsion system design on the carbon footprint of different marine fuels. In: Low Carbon Shipping Conference. London. Balland, O. et al, Optimized selection of vessel air emission controls moving beyond cost-efficiency. Maritime Policy Manage. 42 (4), Bausch, D.O., Brown, G.G., Ronen, D., Scheduling short-term marine transport of bulk products. Maritime Policy Manage. 25 (4), Bengtsson, S., Life cycle assessment of present and future marine fuels. In: Department of Shipping and Marine Technology. Chalmers University of Technology, Gothenburg, p. 71. Bengtsson, S., Fridell, E., Andersson, K., Environmental assessment of two pathways towards the use of biofuels in shipping. Energy Policy 44, Bengtsson, S.K., Fridell, E., Andersson, K.E., Fuels for short sea shipping: a comparative assessment with focus on environmental impact. Proc. Inst. Mech. Eng. Part M: J. Eng. Maritime Environ. 228 (1), Bergqvist, R., Cullinane, K., SECA regulations, modal shift and transport system effects. In: T.A. Logistics and Transport Consultants (Ed.), Gothenburg. Bittner, J., Baird, T., Adams, T., Impacts of Panama Canal expansion on US greenhouse gas emissions. Transport. Res. Rec.: J. Transport. Res. Board 2273, Bond, T.C. et al, Bounding the role of black carbon in the climate system: a scientific assessment. J. Geophys. Res.: Atmosph. 118 (11), Brynolf, S., Fridell, E., Andersson, K., 2014a. Environmental assessment of marine fuels: liquefied natural gas, liquefied biogas, methanol and bio-methanol. J. Clean. Prod. 74, Brynolf, S. et al, 2014b. Compliance possibilities for the future ECA regulations through the use of abatement or change of fuels. Transport. Res. Part D: Transport Environ. 28, Buhaug, Ø. et al, Second IMO GHG Study International Maritime Organization (IMO), London, UK. Carr, E.W., Corbett, J.J., Ship compliance in emission control areas: technology costs and policy instruments. Environ. Sci. Technol. 49 (16), CCNR, Possibilities for Reducing Fuel Consumption and Greenhouse Gas Emissions from Inland Navigation. Report by the Inspection Regulations Committee for the 2012 Autumn Meeting. Annex 2 to protocol 2012-II-4 of the Central Commission for the Navigation of the Rhine. Chatzinikolaou, S., Ventikos, N., Assessment of ship emissions in a life cycle perspective. In: 3rd International Energy, Life Cycle Assessment, and Sustainability Workshop & Symposium (ELCAS3). Nisyros, Greece. Cherubini, F., Bright, R.M., Stromman, A.H., Global climate impacts of forest bioenergy: what, when and how to measure? Environ. Res. Lett. 8 (1), Choi, B.C., Kim, Y.M., 2013a. Theoretical study on fuel savings of marine diesel engine by exhaust-gas heat-recovery system of combined cycle. Trans. Korean Soc. Mech. Eng. B 37 (2), Choi, B.C., Kim, Y.M., 2013b. Thermodynamic analysis of a dual loop heat recovery system with trilateral cycle applied to exhaust gases of internal combustion engine for propulsion of the 6800 TEU container ship. Energy 58, Chryssakis, C., et al., Alternative fuels for shipping, Vol. Position Paper Høvik, Norway, DNV GL. p. 28. Clauss, G., Sickmann, H., Tampier, B., Simulation of the operation of wind-assisted cargo ships. In: 102. Hauptversammlung der Shiffbautechnischen Gesellschaft. Berlin, Germany. CNSS, A Review of Present Technological Solutions for Clean Shipping. Clean North Sea Shipping. Corbett, J.J., Wang, H., Winebrake, J.J., The effectiveness and costs of speed reductions on emissions from international shipping. Transport. Res. Part D: Transport Environ. 14 (8), Cotorcea, A., et al., Present and future of renewable energy sources onboard ships. Case study: solar-thermal systems. Sci. Bull. Mircea cel Batran Naval Acad. 17(1), 35. Cui, Q., Li, Y., An empirical study on the influencing factors of transportation carbon efficiency: evidences from fifteen countries. Appl. Energy 141, Cullinane, S., Mitigating the negative environmental impacts of long haul freight transport. In: Macharis, C., et al. (Eds.), Sustainable Logistics. Emerald Group Publishing Limited, pp Cullinane, K., Bergqvist, R., Emission control areas and their impact on maritime transport. Transport. Res. Part D: Transport Environ. 28, 1 5. Cullinane, K., Khanna, M., Economies of scale in large containerships: optimal size and geographical implications. J. Transport Geogr. 8 (3), Dadd, G.M., Hudson, D.A., Shenoi, R.A., Determination of kite forces using three-dimensional flight trajectories for ship propulsion. Renew. Energy 36 (10), Davidson, M.D., Faber, J., Market-based instruments to reduce greenhouse gas emissions from ships. In: Asariotis, R., Benamara, H., (Eds.), Maritime Transport and the Climate Change Challenge. Abingdon, Earthscan, pp Deniz, C., Thermodynamic and environmental analysis of low-grade waste heat recovery system for a ship power plant. Int. J. Energy Sci. 5 (1), Dnv, G.L., Energy Management Study DNV GL SE, Hamburg. Doulgeris, G. et al, Techno-economic and environmental risk analysis for advanced marine propulsion systems. Appl. Energy 99, Ecoliner, Ecoliner. < [Cited March, 1st 2016]. Eide, M.S. et al, Future cost scenarios for reduction of ship CO 2 emissions. Maritime Policy Manage. 38 (1), Eide, M.S., Chryssakis, C., Endresen, Ø., 2013a. CO 2 abatement potential towards 2050 for shipping, including alternative fuels. Carbon Manage. 4 (3), Eide, M. et al, 2013b. Reducing CO 2 from shipping-do non-co 2 effects matter? Atmosph. Chem. Phys. 13 (8), Einang, P.M., Gas-fuelled ships. In: 25th CIMAC World Congress on Combustion Engine Technology. Vienna. Einang, P.M., New emerging energy sources and systems. In: NorShipping, Lillestrøm. EMEC, Green Ship Technology Book. Brussels, European Marine Equipment Council. Eskeland, G.S., Lindstad, H.E., Environmental taxation of transport. Int. J. Green Growth Dev. 2 (2), Eyring, V. et al, Transport impacts on atmosphere and climate: Shipping. Atmosph. Environ. 44 (37), Faber, J. et al, Technical Support for European Action to Reducing Greenhouse Gas Emissions from International Maritime Transport. CE Delft, NY, pp Faber, J., et al., Going Slow to Reduce Emissions: Can the Current Surplus of Maritime Transport Capacity be Turned into an Opportunity to Reduce GHG Emissions? Seas at Risk, Delft, p. 29. Faber, J., et al., Marginal abatement costs and cost effectiveness of energy-efficiency measures. In: Organization, I.M., (Ed.). The Society of Naval Architects and Marine Engineers (SNAME). London. Fagerholt, K., Ship scheduling with soft time windows: an optimisation based approach. Euro. J. Oper. Res. 131 (3), Fagerholt, K., Johnsen, T.A., Lindstad, H., Fleet deployment in liner shipping: a case study. Maritime Policy Manage. 36 (5), Fagerholt, K., Laporte, G., Norstad, I., Reducing fuel emissions by optimizing speed on shipping routes. J. Oper. Res. Soc. 61 (3), Franc, P., Mitigating maritime transport emissions: diversified effects between European short sea and deep sea shipping. In: Transport Research Arena (TRA) 5th Conference: Transport Solutions from Research to Deployment. Paris, France. Fuglestvedt, J.S. et al, Climate penalty for shifting shipping to the arctic. Environ. Sci. Technol. 48 (22), Future, G.S.O.T., Waste Heat Recovery Systems. In: Future, G.S.O.T. (Ed.). Gilbert, P., From reductionism to systems thinking: how the shipping sector can address sulphur regulation and tackle climate change. Marine Policy 43, Gilbert, P. et al, Technologies for the high seas: meeting the climate challenge. Carbon Manage. 5 (4), Glykas, A., Papaioannou, G., Perissakis, S., Application and cost-benefit analysis of solar hybrid power installation on merchant marine vessels. Ocean Eng. 37 (7), H:\MEPC\72\MEPC 72-INF.12-annex 7

199 420 E.A. Bouman et al. / Transportation Research Part D 52 (2017) MEPC 72/INF.12 Annex 7, page 13 Grahn, M., et al., Cost-effective choices of marine fuels under stringent carbon dioxide targets. In: Proceedings of 3rd International Conference on Technologies, Operations, Logistics and Modelling in Low Carbon Shipping, University College London. Gucwa, M., Schäfer, A., The impact of scale on energy intensity in freight transportation. Transport. Res. Part D: Transport Environ. 23, Guerra, A., Jenssen, M.M., Multi Criteria Decision Analysis (MCDA) in the norwegian maritime sector: adding environmental criteria in maritime decision support systems. In: Faculty of Social Sciences and Technology Management. MSc Dissertation, Norwegian University of Science and Technology, Trondheim, February, 2nd. Halfdanarson, J., Snåre, M.W., Implementation and application of an integrated framework for economic and environmental assessment of maritime transport vessels. In: Department of Energy and Process Engineering. MSc Dissertation, Norwegian University of Science and Technology, Trondheim, Norway. Heitmann, N., Peterson, S., The potential contribution of the shipping sector to an efficient reduction of global carbon dioxide emissions. Environ. Sci. Policy 42, Hertzberg, T., LASS, Lightweight Construction Applications at Sea, SP Report. 2009, SP Technical Research Institute of Sweden: Borås. p Hoffmann, J., Kumar, S., Globalisation the maritime nexus. In: Grammenos, C.T. (Ed.), The Handbook of Maritime Economics and Business. LLP, London, pp Hoffmann, P.N., Eide, M.S., Endresen, Ø., Effect of proposed CO 2 emission reduction scenarios on capital expenditure. Maritime Policy Manage. 39 (4), IPCC, Climate Change, Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. 2007, [Core Writing Team, Pachauri, R.K., Reisinger, A., (Eds.)]. IPCC, Geneva, Switzerland. p Jacobson, M.Z., Short-term effects of controlling fossil-fuel soot, biofuel soot and gases, and methane on climate, arctic ice, and air pollution health. J. Geophys. Res.: Atmosph. 115 (D14). Jafarzadeh, S., Utne, I.B., A framework to bridge the energy efficiency gap in shipping. Energy 69, Jafarzadeh, S., Ellingsen, H., Utne, I.B., Emission reduction in the Norwegian fishing fleet: towards LNG. In: Second International Symposium on Fishing Vessel Energy Efficiency (E-Fishing). Vigo, Spain. Job, S., Why not composites in ships? Reinf. Plast. 59 (2), Johnson, H., Styhre, L., Increased energy efficiency in short sea shipping through decreased time in port. Transport. Res. Part A: Policy Pract. 71, Kesicki, F., Strachan, N., Marginal abatement cost (MAC) curves: confronting theory and practice. Environ. Sci. Policy 14 (8), Kollamthodi, S., et al., Support for the Impact Assessment of a Proposal to Address Maritime Transport Greenhouse Gas Emissions, Report for European Commission DG Climate Action. Ricardo-AEA/R/ED Kontovas, C.A., The Green Ship Routing and Scheduling Problem (GSRSP): a conceptual approach. Transport. Res. Part D: Transport Environ. 31, Kontovas, C.A., Psaraftis, H.N., Transportation emissions: some basics. In: Green Transportation Logistics. Springer, pp Kotb, M., et al., Marine applications of Fuel Cell as alternative power plant: case study. In: Alexandria, E. (Ed.), International Marine and Offshore Engineering Conference (IMOC 2013). Kristensen, H.O., Energy demand and exhaust gas emissions of marine engines. Clean Shipp. Curr. 1 (6), Lack, D., Corbett, J., Black carbon from ships: a review of the effects of ship speed, fuel quality and exhaust gas scrubbing. Atmosph. Chem. Phys. 12 (9), Lauer, A. et al, Global model simulations of the impact of ocean-going ships on aerosols, clouds, and the radiation budget. Atmosph. Chem. Phys. 7 (19), Lee, T.-C., Chang, Y.-T., Lee, P.T., Economy-wide impact analysis of a carbon tax on international container shipping. Transport. Res. Part A: Policy Pract. 58, Lin, S., Greenhouse gas mitigation strategies: a ship operator s perspective in the container shipping industry. In: OAPS (CEE). Nanyang Technological University. Singapore. Lindstad, H., Strategies and measures for reducing maritime CO 2 emissions Ph.D. Dissertation. In: Faculty of Engineering Science and Technology. Norwegian University of Science and Technology, Trondheim. Lindstad, H.E., Hydrogen the next maritime fuel. In: Shipping in Changing Climates Conference Glasgow, UK. Lindstad, H., Eskeland, G.S., Low carbon maritime transport: how speed, size and slenderness amounts to substantial capital energy substitution. Transport. Res. Part D: Transport Environ. 41 (December), Lindstad, H., Sandaas, I., Emission and fuel reduction for offshore support vessels through hybrid technology. In: SNAME Lindstad, H.E., Sandaas, I., Emission and fuel reduction for offshore support vessels through hybrid technology. J. Ship Product. Des. January, 11th Lindstad, H., Asbjørnslett, B.E., Strømman, A.H., Reductions in greenhouse gas emissions and cost by shipping at lower speeds. Energy Policy 39 (6), Lindstad, H., Asbjørnslett, B.E., Strømman, A.H., The importance of economies of scale for reductions in greenhouse gas emissions from shipping. Energy Policy 2012 (46), Lindstad, H., Jullumstrø, E., Sandaas, I., 2013a. Reductions in cost and greenhouse gas emissions with new bulk ship designs enabled by the Panama Canal expansion. Energy Policy 59, Lindstad, H., Asbjørnslett, B.E., Jullumstrø, E., 2013b. Assessment of profit, cost and emissions by varying speed as a function of sea conditions and freight market. Transport. Res. Part D: Transport Environ. 19, Lindstad, H., Sandaas, I., Steen, S., Assessment of profit, cost, and emissions for slender bulk vessel designs. Transport. Res. Part D: Transport Environ (29), Lindstad, H., et al., GHG Emission Reduction Potential of EU-Related Maritime Transport and on its Impacts. European Commission, CLIMA.B.3/ETU/ 2013/0015, TNO 2014 R Lindstad, H. et al, 2015b. Maritime shipping and emissions: a three-layered, damage-based approach. Ocean Eng. 110 (Part B), Lindstad, H., Sandaas, I., Strømman, A.H., 2015c. Assessment of cost as a function of abatement options in maritime emission control areas. Transport. Res. Part D: Transport Environ (38), Lindstad, H., Bright, R.M., Strømman, A.H., 2016a. Economic savings linked to future Arctic shipping trade are at odds with climate change mitigation. Transport Policy 45, Lindstad, H., Asbjørnslett, B.E., Strømman, A.H., 2016b. Opportunities for increased profit and reduced cost and emissions by service differentiation within container liner shipping. Maritime Policy Manage. 43 (3), Lock, L.M., Future Fuel for Worldwide Tankershipping in Spot Market MSc Dissertation. KTH Royal Institute of Technology, Stockholm, Sweden. Ludvigsen, K.B., Ovrum, E., Fuel Cells for Ships, Høvik, Norway, Det Norske Veritas. Ma, H. et al, Well-to-wake energy and greenhouse gas analysis of SOX abatement options for the marine industry. Transport. Res. Part D: Transport Environ. 17 (4), Maddox Consulting, Analysis of Market Barriers to Cost Effective GHG Emission Reductions in the Maritime Transport Sector. CLIMA.B.3/SER/2011/ 0014, London. Mastrandrea, M.D., et al., Guidance Note for Lead Authors of the IPCC Fifth Assessment Report on Consistent Treatment of Uncertainties. Intergovernmental Panel on Climate Change (IPCC). < McCord, M.R., Lee, Y.-K., Lo, H.K., Ship routing through altimetry-derived ocean currents. Transport. Sci. 33 (1), H:\MEPC\72\MEPC 72-INF.12-annex 7

200 Annex 7, page 14 E.A. Bouman et al. / Transportation Research Part D 52 (2017) Miola, A., Marra, M., Ciuffo, B., Designing a climate change policy for the international maritime transport sector: market-based measures and technological options for global and regional policy actions. Energy Policy 39 (9), Myhre, G. et al, Anthropogenic and natural radiative forcing. In: Stocker, T.F. et al. (Eds.), Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, pp Nikolakaki, G., Economic incentives for maritime shipping relating to climate protection. WMU J. Maritime Aff. 12 (1), Norlund, E.K., Gribkovskaia, I., Reducing emissions through speed optimization in supply vessel operations. Transport. Res. Part D: Transport Environ. 23, Norstad, I., Fagerholt, K., Laporte, G., Tramp ship routing and scheduling with speed optimization. Transport. Res. Part C: Emerg. Technol. 19 (5), Nuttall, P.R., Sailing for Sustainability: The Potential of Sail Technology as an Adaptation Tool for Oceania. A Voyage of Inquiry and Interrogation Through the Lens of a Fijian Case Study. Ph.D. Dissertation, Victoria University of Wellington. Pauli, G., Emissions and inland navigation. In: Green Transportation Logistics. Springer, pp Paxian, A. et al, Present-day and future global bottom-up ship emission inventories including polar routes. Environ. Sci. Technol. 44 (4), Pélerin, E., Tincelin, T., EOSEAS green cruise ship concept. In: 7th Annual Green Ship Technology Conference. Copenhagen, March, 16th. Poulsen, R.T., Sornn-Friese, H., Achieving energy efficient ship operations under third party management: how do ship management models influence energy efficiency? Res. Transport. Bus. Manage. 17, Psaraftis, H.N., Market-based measures for greenhouse gas emissions from ships: a review. WMU J. Maritime Aff. 11 (2), Psaraftis, H.N., Green maritime transportation: market based measures. In: Psaraftis, H.N. (Ed.), Green Transportation Logistics. Springer International Publishing, Switzerland, pp Psaraftis, H.N., Kontovas, C.A., Speed models for energy-efficient maritime transportation: a taxonomy and survey. Transport. Res. Part C: Emerg. Technol. 26, Qiu, Y., Sun, Y., Yuan, C., Review on the application and research progress of photovoltaics-ship power system. In: The 3rd International Conference on Transportation Information and Safety. IEEE, Wuhan, PR China. Ranheim, E., Hallet, G., Virtual Arrival A Way to Reduce Greenhouse Gas (GHG) Emissions. INTERTANKO and OCIMF, ER-17652/ Rialland, A. et al, Performance-based ship management contracts using the Shipping KPI standard. WMU J. Maritime Aff. 13 (2), Rojon, I., Dieperink, C., Blowin in the wind? Drivers and barriers for the uptake of wind propulsion in international shipping. Energy Policy 67, Russell, B.A., Wang, H., Zeinali, M., Marginal abatement costs of CO 2 emissions reduction and market-based mechanisms and the pricing of credits. In: Climate Change and Ships: Increasing Energy Efficiency. Linthicum H., Maryland, February, Sánchez-Heres, L.F., Opportunities for Weight Reduction in Composite Marine Structures. Ph.D. Dissertation, Chalmers University of Technology, Gothenburg, Sweden. Schmitz, M., Madlener, R., Economic viability of kite-based wind energy powerships with CAES or hydrogen storage. Energy Proc. 75, Sciberras, E.A. et al, Electric auxiliary propulsion for improved fuel efficiency and reduced emissions. Proc. Inst. Mech. Eng. Part M: J. Eng. Maritime Environ. 229 (1), Seddiek, I., An overview: environmental and economic strategies for improving quality of ships exhaust gases. Trans. RINA, Int. J. Marit Eng. 157 (Part A1), Shipping & Marine, Breakthrough in Composite Approval. Schofield Publishing Ltd. Sjöbom, K., Magnus, P., Energieffektivisering ombord M/S Sydfart: Med hjälp av solceller. In: Fakulteten För Teknik. Sjöingenjörsprogrammet, Uppsala, Sweden. SkySails, SkySails Propulsion System. < [March, 1st 2016]. Smith, T., et al., Analysis techniques for evaluating the fuel savings associated with wind assistance. In: Low Carbon Shipping Conference. London. Smith, T. et al, Third IMO GHG Study International Maritime Organization (IMO), London. Solem, S. et al, Optimization of diesel electric machinery system configuration in conceptual ship design. J. Marine Sci. Technol. 20 (3), Stott, P., Wright, P., Opportunities for improved efficiency and reduced CO 2 emissions in dry bulk shipping stemming from the relaxation of the Panamax beam constraint. Int. J. Maritime Eng. 153 (A4), A215 A229. Styhre, L., Capacity utilisation in short sea shipping. In: Department of Technology Management and Economics. Ph.D. Dissertation, Chalmers University of Technology, Gothenburg. Sys, C. et al, In search of the link between ship size and operations. Transport. Plan. Technol. 31 (4), Taljegard, M. et al, Cost-effective choices of marine fuels in a carbon-constrained world: results from a global energy model. Environ. Sci. Technol. 48 (21), Teeter, J.L., Cleary, S.A., Decentralized oceans: sail-solar shipping for sustainable development in SIDS. In: Natural Resources Forum. Wiley Online Library, pp Thomson, H., Corbett, J.J., Winebrake, J.J., Natural gas as a marine fuel. Energy Policy 87, Tillig, F., Mao, W., Ringsberg, J., Systems Modelling for Energy-Efficient Shipping. Chalmers University of Technology. Traut, M. et al, Propulsive power contribution of a kite and a Flettner rotor on selected shipping routes. Appl. Energy 113, Ulstein (Ed.), ULSTEIN X-BOW Ò - A Vision Turned into Reality. Ulstein Design AS, Ulsteinvik, Norway. Wang, H., Lutsey, N., Long-term potential for increased shipping efficiency through the adoption of industry-leading practices. In: International Council on Clean Transportation, September 30th. Wang, H., et al., Marginal abatement costs and cost effectiveness of energy-efficiency measures. In: Organization, I.M. (Ed.), The Society of Naval Architects and Marine Engineers (SNAME). London. Wärtsila, Boosting energy efficiency: energyefficiency catalogue. In: Energy Efficiency Catalogue/Ship Power R&D. Wärtsila. Welaya, Y., Mosleh, M., Ammar, N.R., Thermodynamic analysis of a combined gas turbine power plant with a solid oxide fuel cell for marine applications. Int. J. Naval Architect. Ocean Eng. 5 (4), Woo, J.-K., Moon, D.S.-H., The effects of slow steaming on the environmental performance in liner shipping. Maritime Policy Manage. 41 (2), Wu, W.-M., Lin, J.-R., Productivity growth, scale economies, ship size economies and technical progress for the container shipping industry in Taiwan. Transport. Res. Part E: Logist. Transport. Rev. 73, Yin, J. et al, Slow steaming of liner trade: its economic and environmental impacts. Maritime Policy Manage. 41 (2), Zöllner, J., Strömungstechnische Möglichkeiten zur Reduzierung des Kraftstoffverbrauchs und der CO 2 -Emissionen von Binnenschiffen. Vortrag beim ZKR Kongress Rheinschifffahrt und Klimawandel. Bonn, June 24 25th. H:\MEPC\72\MEPC 72-INF.12-annex 7

201 Annex 8, Page 1 ANNEX 8 CONTRIBUTION TO REQUEST 4 AND 5 OF THE FIRST ROUND BY RUSSIAN FEDERATION Correspondence Group on EEDI Review beyond phase 2 According working schedule of the Correspondence Group Russian Federation was sponsor of documents MEPC 71/5/2 and MEPC 71/ INF.7, which should be considered by correspondent group. Below we present our position taking into consideration the aspects of discussion held on MEPC71. Our documents MEPC 71/5/2 and MEPC 71/ INF.7 raised 3 issues, correspondingly our positions on each of them are formulated in items 1-3 of present paper. 1 Implementation of 5% margin to basic lines for all ice class ships. This issue was raised because it was found that due to safety reasons it is impossible to apply a great majority of energy saving devices (ESD) on ice class ships. For the ships without ice class ESD is one of the main measures to meet requirement on power reduction prescribed for Stage 1 of EEDI regulation. That is why at each stage of EEDI regulation the ice class ships will have lower potential for power reduction than the ships without ice class. The value 5% can be justified by the analysis of ESD application effect, presented on 21 ITTC held in 2014, Copenhagen. The results of this analysis were published in Proceeding of the Conference which is in free access on ITTC website (ittc.inf) in item Report of Propulsion Committee. The analysis was based on a great number of experimental studies carried out in towing basins worldwide in support of ESD design (in practice all ESD are designed with tuning in towing tank model tests). For analysis all ESD were divided into 3 groups: А. ESD in front of propeller; В. ESD after propeller; С. ESD on the propeller. А. Group of ESD in front of propeller includes different pre-ducts and foil systems for pre- swirling or equalizing propeller inflow, improving propeller operation conditions. In the ITTC propulsion committee report (Table 4) it was found that the efficiency of ESD for the ships tested in towing basins, as a rule, was in the range 3-5%. The highest efficiency was demonstrated by Mewis Duct, which combines the effects of equalizing duct and pre-swirl devices (due to special foils inside the duct). Efficiency of such ESD as a rule varied in the range 4-7.7% (table 6 of ITTC Propusion Committee report). One should take into consideration that all such ESD demonstrate different effect of power reduction in full load and ballast load conditions even on the same ship.

202 Annex 8, Page 2 Shipowners do not accept all such ESD for the ships of ice class because of the danger that during contact with ice ESD will be torn off the hull probably together with a portion of hull plating. В. ESD after propeller usually are the post swirl foils system installed after propeller for utilize swirling effect in propeller slipstream. One of the examples is Grim wheel with efficiency of power reduction estimated at 6%. But such devices have a strength problem even on the ships without ice class because they permanently work in conditions of fluctuating loading in propeller slipstream. In ice conditions the strength problem for such devices looks too difficult to be resolved. Also, ESD installed after propeller hub - PBCF (propeller Cone Fin) should be mentioned. It is a compact foil system operating in the hub vortex and recovering the energy of this vortex. Efficiency of such ESD was registered to be within 3%. In principle they can be applied for ships of low ice class with the propeller installed as traditional on the shaft. But the ships of high and medium ice classes are now more frequently outfitted with podded propulsion systems with pulling propellers. In this case the propeller is installed in front of pod and so application of PBCF is impossible. Also, there is a problem with PBCF strength at reversing in ice, because in this case foils of PBCF will operate in front of propeller in direct contact with ice. This ESD group also includes different foil systems installed on the rudder. Their efficiency is usually up to 3% (Table 3 of ITTC Propulsion Committee report), but it is unacceptable for ice class ships due to possible interaction with ice and danger of accident with steering gear. Application of asymmetrical rudders with post swirling effect on non-ice-class ships were also reported.. In ITTC report, the power reduction effect of these systems is estimated at up to 4-6%. But until now the asymmetrical rudder has been tested only for single screw ships, so the ITTC conclusion is that at present there is no sufficient information at present to justify practical implementation of this solution. The problems related to application of asymmetric rudders in ice have not been considered at all until now. However, it is clear, that asymmetrical rudder may cause asymmetry of forces on the rudder in ice and it means significant change of requirements for the steering gear characteristics. Also side force on asymmetrical rudders may be very unfavorable when the ship is maneuvering astern in ice. С. Regarding the ESD on propeller, the ITTC Proceeding considered application of new type of propellers with tips of blades bent forward or aft - CLT and DTV (Cappel s propeller). The energy saving effect of such propellers according to ITTC estimation and model tests of Krylov Centre, Russia, is up to 5%. But in Russia numerous cases were reported, when even blades of traditional skewed propellers were bent at contact with ice. Application of artificially strongly bent blades of CLT and DTV exclude the ice milling mode, which is the most important operating mode for ice class propellers, because such blade will inevitably be broken at contact with ice.. Thus, the ITTC materials, being a reflection of the world shipbuilding experience with ESD design, demonstrate that the maximum effect of ESD is about 6-7%

203 Annex 8, Page 3 (as a rule it refers to ESD taking the benefit of two combined energy-recovering effects for example pre-swirl plus wake equalizing effects). The overwhelming majority of ESD is not suitable for ice class ships due to safety reasons. An exception is PBCF installed after propeller hub for recovery of hub vortex energy, but the energy saving effect of PBCF is 2-3% with an additional strength problem to be addressed separately. Also for ice class propellers the blades are designed with verification of strength for operation in contact with ice (milling mode). So, blades are always thicker than for propellers without ice class and it leads to some reduction of propeller efficiency. Considering that application of ESD on ice class ship is impossible in general, it is reasonable to decrease the required power reduction by 5% for ice class ships at each stage of EEDI implementation in view of the fact that according to ITTC data this value of 5 % is the averaged effect of power reduction due to ESD application. 2 The minimum power for ice-class ships IMO is currently developing guidelines for determining minimum propulsion power to maintain the capability of navigating and safe maneuvering in adverse conditions for ships without ice class. The Russian Federation notes that classification societies have introduced minimum power requirements for all ice-class ships based on the safety of navigation in ice, depending on the ice limits for different ice classes and ice conditions in which the ship is supposed to be operated. In the Annex, excerpts from Rules of Russian Maritime Register of Shipping are given as an example. The Finnish-Swedish Rules also specify requirements for minimum power of ice ships. The only point of discussion with respect to the implementation of this proposal is the lack of minimum power requirements in the IACS classification. However, considering that there is a generally recognized table of ice class equivalence for the ice classes IA Super and lower in the HELCOM recommendation 25/7, the minimum power requirements for IA Super and IA can be applied to the minimum power of IACS PC6 and PC7 respectively. The higher ice classes PC1 to PC5 or equivalent that are not included in the HELCOM table should not be included in the EEDI calculation methodology as discussed at MEPC 71 and according to the instructions given to the correspondence group. In any case, the designs of ice-class ships should comply with the rules of classification societies, that is, ice class ships should have onboard the minimum power conforming to the specified rules with due account for the ice class. In this regard, our proposal is that for ice ships the minimum power should be determined using two methods - according to the guidelines developed by IMO for ships without ice classes and according to the formulas for determining the minimum power depending on ice class, and take as the largest one of the two values. Without implementing such a proposal, the IMO-developed methodology for determining the minimum power will be incomplete for ice ships.

204 Annex 8, Page 4 It is expected that the minimum power values calculated according to the two methods will be close to the low ice classes, while for the medium and high ice classes the power determined by the formulas for ice ships will be substantially higher. Since there are no minimum power requirements in the IACS classification, for classes PC7 and PC6 it is possible to apply the power requirement of the equivalent IA Super and IA classes according to the HELCOM recommendation table. For the PC5 PC1 classes, since they are considered "higher" than the IA Super, we offer not apply to energy efficiency requirements. We believe that the correspondence group should recommend MEPC to include a proposal for taking into account the minimum power required for the ice class in question into the general IMO guidelines for determining the minimum propulsion power. 3 The third proposal of the Russian Federation in the document 71/2 was not to apply the EEDI requirements to the ships of the ice classes hither than 1A Super. The main point of discussion in this respect is what ice classes should be taken out of the regulation, i.e., a question regarding the table of correspondence between ice classes of different classification systems and the classification of IACS is raised. One of the major arguments in favor of not applying EEDI regulation to the ships with high ice classes is the presence on board of a large power reserve that is used for ice operations only, while for the movement in the free water less than a half capacity is typically used. Dependence of the safety of ships navigating in the polar waters on the main engine power can be demonstrated with the following possible situations: - A ship with a lack of power may be captured by the drifting ice, and it may become damaged due to grounding or collision with underwater rocks; - A ship captured by ice due to the insufficient power can be trapped in ice for the entire winter period, being exposed to the danger of being crushed due to the ice compression, - In heavy ice conditions, the lack of power may cause the propeller jamming, engine stop and loss of stroke that could result in the damage to the propeller or hull elements. In all these cases, rescue operations will be required to free the ship and evacuate the people. The power required by the classification societies for safe ice navigation does not allow the ships to meet the requirements of MARPOL Annex VI Chapter 4, since the required power is significantly higher than the level allowed for compliance with the EEDI performance criterion. Table 1 of MEPC.212 (63) provides correction factors for the ships with ice class IA Super, IA, IB and IC and equivalent according to the HELCOM Recommendation 25/7, i.e. the classes that are qualified as categories B and C in terms of strength in the Polar Code. However, this table does not take into account the ice ships designed for operation in multi-year ice that are qualified as category A in the Polar Code. We believe that the requirements of MARPOL Annex VI Chapter 4 should not apply to the category A ships (PC1-PC5 and equivalent) the same as provided for icebreakers. We suggest that this proposal should be submitted for discussion to the Marine Environment Protection Committee (MEPC). Below is a table demonstrating an approximate correspondence of classes and specifying the classes to be considered "above 1A Super". This table is based on the table of equivalence of the Baltic classes and ice classes of different classification societies. In the second and third columns, the classes are taken from the annex to the HELCOM Recommendation 25/7 with the exception of those classes that are not currently used for new construction (since the

205 Annex 8, Page 5 requirements for the EEDI are not applicable to existing ships). In the fourth column the classes «higher than IA Super» and not included in the HELCOM Recommendation 25/7 are given. The authors would be grateful to the members of the correspondence group and to the representatives of the classification societies for checking the correctness of the classes indicated by us. Any additions and comments will be highly appreciated. It would be extremely useful if the representatives of the classification societies removed those classes that are not currently used for new construction. Table of correspondence between ice classes of different classification societies Classification Ice Class Society Finnish-Swedish Class Rules Ice Russian Maritime Register of Shipping (Rules 2008) American Bureau of Shipping Bureau Veritas CASPPR, 1972 China Society Classification Det Norske Veritas DNV GL Germanischer Lloyd Ice Classes lower than IA Super IА IB IC Category II Arc 4 Ice 3 Ice 2 Ice 1 Ice Class I А Ice Class I B Ice Class I C D0 ICE CLASS IА ICE CLASS IВ ICE CLASS IC ID В С D E Ice Class B1 Ice Class B2 Ice Class B3 Ice Class B ICE-1А ICE-1B ICE-1С ICE-C Ice(1A) Ice(1B) Ice(1C) - Е3 E2 E1 E Ice Classes equal to IA Super Ice classes higher than IA Super IА Super - Arc 5 Arc 6, Arc 7, Arc 8, Arc 9, Icebreaker 6, Icebreaker 7, Icebreaker 8, Icebreaker 9 Ice Class I АА А0, A1, A2, A3, A4, A5 ICE CLASS IA SUPER А - CAC4, CAC3, CAC2, CAC1 Ice Class B1* - ICE-1А* ICE-05, ICE-10, ICE-15, POLAR-10, POLAR-20, POLAR- 30 Ice(1A*) - Е4 Arc1 - Arc4 IACS Polar Rules PC7 PC6 PC5, PC4, PC3,PC2, РС1

206 Annex 8, Page 6 Korean Register of Shipping Lloyd s Register of Shipping Nippon Kaiji Kyokai PolskiRejestrStatków RegistroItalianoNavale IA IB IC ID Ice Class 1A FS (+) Ice Class 1A FS Ice Class 1B FS (+) Ice Class 1B FS Ice Class 1C FS (+) Ice Class 1C FS Ice Class 1D Ice Class 1E NS* (Class IA Ice Strengthening) NS (Class IA Ice Strengthening) NS* (Class IB Ice Strengthening) NS (Class IB Ice Strengthening) NS* (Class IC Ice Strengthening) NS (Class IC Ice Strengthening) NS* (Class ID Ice Strengthening) NS (Class ID Ice Strengthening) L1 L2 L3 L4 ICE CLASS IA ICE CLASS IB ICE CLASS IC ID IA Super - Ice Class 1AS FS (+) Ice Class 1АS FS NS* (Class IA Super Ice Strengthening) NS (Class IA Super Ice Strengthening) Ice Class AC3 Ice Class AC2 Ice Class AC1.5 Ice Class AC 1 L1A - ICE CLASS IA SUPER - -

207 Annex 8, Page 7 ANNEX Explanation of the requirements of the Register Rules for the minimum power of ice class ship. 1 Operational modes of ice ship s powerplants Traditional displacement transport ships, as a rule, operate at speeds close to the design full speed. Transport ice-strengthened ships and icebreakers rarely operate at a full speed which their powerplant is capable to provide in open water. With the propulsive output determined for extreme operational conditions, the ships in question have a large power reserve. Ice navigation imposes special requirements upon the ship s machinery installation. The right choice of plant components essentially affects the efficient and accident-free operation of the ship. The operation of the ship s propulsion plant in ice features an opportunity of variations of ship s resistance and the screw rotation resisting moment within a wide range in a moderate period of time. Thus one of the basic requirements placed upon ice ship propulsion plants is a necessity to maintain the propulsion plant output required within the entire range of the variation of engine operational modes, i.e. from mooring conditions to ice-free navigation. In addition, operational ice conditions demand high manoeubrability, i.e. an opportunity to frequently change modes and a rotation direction. The specific mode of propulsion plant operation in ice is associated with the screwice interaction. This mode is the heaviest both for the screw and the propulsion plant at large. With blade-ice impacts, the screw rotation resisting moment drastically increases resulting in the fast change of speed and, in some cases, in the full stop of the propeller shaft (screw jamming) despite the significant propeller shaft torque available. The full stop of a screw navigating in ice is very risky as, in this case, the blade tows oncoming ice-floes without their breaking and throwing away. If the blade-towed ice-floe therewith sets against the ice channel edge, the forces taken by the blade may damage the screw or propeller shaft. To prevent screw jamming, it is desirable during design to provide the propulsion plant with the propeller torque exceeding the moment necessary for ice destruction. Thus, the basic additional requirements to be taken into account while making a choice of the powerplant type for the ship operating in ice, may be settled:.1 suddenly change of loading shall not result in reduction of propeller output and torque;.2 with reduction of the propeller speed down to a stop, high shaft loading (torque) is to be sustained;.3 high manoeuvrability, i.e. an opportunity of frequent and quick reverses. So, ice ships powerplants feature is the volatility of the loading and a large power reserve during most of the running time. Therefore, the Rules of the Russian Maritime Register of Shipping provide for minimum power requirements (see of part VII RS Rules). In of part VII RS Rules

208 Annex 8, Page 8 there are given alternative requirements to the power of vessels (same as requirements of the Finnish-Swedish Rules) that can be used for low ice classes Ice2, Ice3 and Arc4. The minimum required power propeller shaft of ships of ice classes Arc5 to Arc9 shall be determined according to only. 2 RS Requirements for minimum power output. Based on operational modes of icebreakers and ice ships, the Register s Rules set the minimum total shaft power necessary for the assignment of a particular class. For Ice2 to Ice3 and Arc4 to Arc 9 ice category ships, the minimum power shall be determined by the formula: P min f 1 f 2 f 3 ( f P0) 4 where Δ ship s displacement to a summer load waterline, and coefficients f 1, f 2, f 3 take into account ship s various data affecting speed performance in ice: ship s breadth, presence of a CPP or use of electric propulsion. Depending on the value of the factors listed, the product f 1f 2f 3 varies within 0,85 to 1,1. f 1 =1,0 for fixed pitch propellers; f 1 = 0,9 for propulsion plants with controllable pitch propellers or electric drive; f 2 = φ / ,675 but not more than 1,1; φ = slope of stem (refer to , Part II "Hull" - Fig. 1 ); f 2 =1,1 for a bulbous stem; the product f 1 f 2 shall be taken in all cases not less than 0,85; 3 f 3 = 1,2B/ Δ but not less than 1,0; B = breadth of the ship, m Δ = ship's displacement to the summer load waterline (refer to 1.2.1, Part III "Equipment, Arrangements and Outfit"), t. Fig. 1 φ slope of stem at the level of the summer cargo waterline. The values of P 0 and f 4 are determined by the table given in the Rules depending on the ice category ( see T a b l e ).

209 Required power, МВт MEPC 72/INF.12 Annex 8, Page 9 T a b l e Displacement Value Ice class ships Ice2 Ice3 Arc4 Arc5 Arc6 Arc7 Arc8 Arc9 Δ < f 4 0,18 0,22 0,26 0,3 0,36 0,42 0,47 0,5 P 0, kw Δ f 4 0,11 0,13 0,15 0,2 0,22 0,24 0,25 0,26 P 0, kw If the product of those three coefficients is taken equal to 1 as the first approximation, the power required by the Register s Rules depending on the ship s displacement and ice category is presented in Fig Displacement Ice1, Ic Ice3 Arc4 Arc5 Arc6 Arc7 Arc8 Arc9 Fig. 2. Register s Rules requirements for the minumum power of ice-strengthened ships depending on displacement (assumed f 1f 2f 3 = 1). Irrespective of the results obtained in calculating the power as per above formula, the minimum power, in kw, shall not be less than (see Fig. 3): kw for Arc9 ice class ships; 7200 kw for Arc8 ice class ships; 5000 kw for Arc7 ice class ships; 3500 kw for Arc6 ice class ships; 2600 kw for Arc5 ice class ships; 1000 kw for Arc4 ice class ships; 740 kw for ships of ice classes Ice2 and Ice3.

210 Minimum reguired power, МВт MEPC 72/INF.12 Annex 8, Page Ice2 Ice3 Arc4 Arc5 Arc6 Arc7 Arc8 Arc9 Fig. 3. Register s Rules requirements for the minimum power of ice-strengthened ships. ***

211 Annex 9, Page 1 ANNEX 9 CONTRIBUTION TO REQUEST 1 OF THE SECOND ROUND BY JAPAN Japan would like to provide the outcome of case study for bulkers and tankers as follows: Fig. 1 and 2 show the best possible attained EEDI when applying all the practically available energy saving to the latest energy efficient ships. Fig. 3 shows the highest achievable energy saving rate for each size of bulkers and tankers based on the result of the study. 10 Bulk carrier Attained EEDI Attained EEDI Ref. Line 10% 20% 30% , , ,000 DWT Fig. 1 Attained EEDI when applying all the practical energy saving (bulkers) 14 Tanker Attained EEDI Ref. Line 10% 20% 30% Attained EEDI , , ,000 DWT Fig. 2 Attained EEDI when applying all the practical energy saving (tankers)

212 Kind of Ship MEPC 72/INF.12 Annex 9, Page 2 Bulk carrier - Handy Bulk carrier - Handymax Case 1 Case 2 Bulk carrier - Panamax Bulk carrier - Cape Bulk carrier - VLOC Further measures such as MCR reduction are needed. Tanker - small Tanker - Aframax Tanker - VLCC Reduction rate with all effective ESDs [%] Fig. 3 Achievable reduction rate for bulkers and tankers Case-1 applies pre/post-swirl devices, air lubrication system or low friction coating, waste heat recovery system and solar power Case-2 applies coaxial contra-rotating propeller, air lubrication system or low friction coating, waste heat recovery system and solar power The outcome of this study shows the following points: - When all the practical energy saving are applied to the latest bulkers and tankers, most of the EEDI reduction rate will be around 25 percent and few ships can meet 30 percent reduction rate. - In order for these ships to meet 30 percent reduction rate of attained EEDI, it is necessary to use the engine with smaller MCR, which leads to lower design speed, or to apply additional energy saving, which are not generally used because of the limitation of applicability (e.g. LNG fuel which requires supply system of LNG on shore). In addition, we need to bear in mind that the potential of improving energy efficiency by reducing MCR is limited because minimum power requirements are defined in MARPOL and related guidelines. In this context, setting unnecessarily stringent minimum power requirement will be a barrier to GHG reduction from new ships. For the other ship types, it is necessary to make further analysis. In general, hydrodynamic energy saving have less effects on the performance of fast ships than that of slower ships such as bulkers and tankers because fast ships have slender hull form with better hydrodynamic performance of which potentials for further energy efficiency improvement through hydrodynamic energy saving is limited. ***

213 Annex 10, page 1 SJØFARTSDIREKTORATET EEDI Assessment of for reduced EEDI Sjøfartsdirektoratet Report No.: , Rev. 0 Document No.: 1175HHZT-3 Date: H:\MEPC\72\MEPC 72-INF.12-annex 10

214 Annex 10, page 2 Project name: Sjøfartsdirektoratet EEDI DNV GL AS Maritime Report title: Assessment of for reduced EEDI Shipping Advisory Customer: Sjøfartsdirektoratet, Postboks HAUGESUND Norway Veritasveien Høvik Norway Tel: Customer contact: Date of issue: Project No.: Organisation unit: Shipping Advisory Report No.: , Rev. 0 Document No.: 1175HHZT-3 Applicable contract(s) governing the provision of this Report: 1-175MYFR-MRNNO729-1 Objective: Sjøfartsdirektoratet has asked DNV GL to perform case studies according to point 2.1 of document Correspondence Group on EEDI Review beyond phase 2 Round 2 dated 6 th November 2017 Focus shall be on the vessel segments (type and size) for which the current EEDI requirements could be made more strict or subject for earlier EEDI Phase 3 implementation The segments to be evaluated will be agreed at project startup The results will be used as input to the Correspondence Group (CG) and shall be delivered before the next CG session on the 11 th of December Prepared by: Verified by: Approved by: Liv Hagen Consultant Eivind Ruth Principal spcialist Knut Ljungberg Head of Section [Name] [title] [Name] [title] [Name] [title] [Name] [title] DNV GL Report No , Rev. 0 H:\MEPC\72\MEPC 72-INF.12-annex 10 Page i

215 Annex 10, page 3 Copyright DNV GL All rights reserved. Unless otherwise agreed in writing: (i) This publication or parts thereof may not be copied, reproduced or transmitted in any form, or by any means, whether digitally or otherwise; (ii) The content of this publication shall be kept confidential by the customer; (iii) No third party may rely on its contents; and (iv) DNV GL undertakes no duty of care toward any third party. Reference to part of this publication which may lead to misinterpretation is prohibited. DNV GL and the Horizon Graphic are trademarks of DNV GL AS. DNV GL Distribution: Unrestricted distribution (internal and external) Unrestricted distribution within DNV GL Group Unrestricted distribution within DNV GL contracting party No distribution (confidential) Keywords: EEDI, energy efficiency, IMO Rev. No. Date Reason for Issue Prepared by Verified by Approved by First issue Hagen, Liv Aune Ruth, Eivind Ljungberg, Knut DNV GL Report No , Rev. 0 H:\MEPC\72\MEPC 72-INF.12-annex 10 Page ii

216 Annex 10, page 4 Table of contents 1 BASIS FOR WORK SELECTION OF RELEVANT ENERGY SAVING DEVICES EVALUATION OF MINIMUM PROPULSION POWER AND THE SHIP EFFICIENCY IMPROVEMENT POTENTIAL FOR TANKERS AND BULK CARRIERS GENERAL CARGO SHIP EFFICIENCY IMPROVEMENT POTENTIAL SUMMARY OF EEDI SCORE POTENTIALS CONSIDERATIONS ON REQUEST REFERENCES DNV GL Report No , Rev. 0 H:\MEPC\72\MEPC 72-INF.12-annex 10 Page iii

217 Annex 10, page 5 1 BASIS FOR WORK The Norwegian Maritime Authority has requested DNV GL to perform case studies according to point 2.1 of document Correspondence Group on EEDI Review beyond phase 2 Round 2 dated November 6 th The focus is to be on the vessel segments (type and size) for which the current EEDI requirements could be made stricter or subject for earlier EEDI phase 3 implementation. The vessel segments to study were selected to cover a large span of ship types and sizes with focus on possibly earlier or more stringent EEDI phase 3 implementation. For each segment, a set of reference ships were selected. To evaluate the potential for improvement in efficiency, the vessels with the best EEDI scores were chosen for reference in each of the four categories. This resulted in vessels built and set in operation within the last few years. DNV GL classed vessels were used as reference as this gave us the most information on each individual ship. To maintain the anonymity of the vessels they are presented using intervals of DWT capacity. The following ship types and sizes were chosen: VLCC tanker ( DWT), EEDI score % below phase 0 level Panamax bulk carrier ( DWT), EEDI score % below phase 0 level Chemical/product tanker ( DWT), EEDI score % below phase 0 level General cargo ship ( DWT), EEDI score % below phase 0 level DNV GL Report No , Rev. 0 Page 1 H:\MEPC\72\MEPC 72-INF.12-annex 10

218 Annex 10, page 6 2 SELECTION OF RELEVANT ENERGY SAVING DEVICES To estimate the efficiency improvement potentials of the VLCC tanker, Panamax bulk carrier, chemical/product tanker and general cargo ships fleets, relevant energy saving and their saving potentials have been identified. Only the measures that are rarely/sometimes used are considered, as the that are commonly used are most probably already implemented on the reference vessels. The relevance and reduction potentials of each technology for the different ship types are presented in Table 2-1. The relevant energy saving are taken from /1/ (which is conducted from comments of the IMO member country delegates latest input on different energy saving devices) and supplemented by DNV GL experience. In the same document (/1/), the reduction potentials are in many cases specified in percentage intervals and in some cases, they are only relevant for certain ship types and sizes. The presented reduction potentials are revised and altered, based on DNV GL experience, to better comply with each ship type and size. The reduction potentials are meant to quantify the reduction in power use that a typical ship for the specific ship types can expect to achieve when employing the different, and can thus be slightly unconservative when evaluating the best EEDI performers in each category. In many cases a combination of results in a smaller improvement than the isolated effect of one of the measures. E.g. in the isolated effects of pre- and post-swirl devices are estimated to 1-4 % for each measure, but in combination the total improvement ratio is predicted to be 4-6 % (/1/). Table 2-1 Relevant energy saving devices for each ship type Relevant energy Panamax Chemical/pr # VLCC tanker saving bulk carrier oduct tanker 1 Podded (azipod) drives General cargo ship 0 % 0 % 0 % 0 % Comments Not effective for other ship types than tugs, supply vessels and other vessels for short sea shipping Coaxial contrarotating propeller Free rotating vane wheel Ducted propeller Unconventiona l propeller Pre-swirl devices 4 % 4 % 5 % 5 % 3 % 3 % 4 % 4 % 4 % 4 % 0 % 0 % 2 % 2 % 2 % 2 % 4 % 4 % 3 % 3 % Post-swirl devices 1 % 1 % 2 % 2 % The total reduction potential for related to recovery of propeller energy (2-7) is estimated to be 6 % for the VLCC and Panamax bulk carrier (assuming a combination of pre- and post-swirl devices and a ducted DNV GL Report No , Rev. 0 Page 2 H:\MEPC\72\MEPC 72-INF.12-annex 10

219 Annex 10, page Split stern (twin skeg) Machinery system 0 % 0 % 0 % 0 % 2 % 2 % 2 % 2 % 10 PTO/PTI 1 % 1 % 1 % 2 % propeller). For the chemical/pro duct tanker the reduction potential is estimated to 6 % regardless of which achievable combination is used. Relevant for ultra-wide ships. A collective category for optimization of the machinery system. Tuning of the main engine wrt. operation as opposed to a design point is considered the most effective measure within the category. The potential for a reduction in CO2 emissions is deemed larger for the two smaller ships, but it will not affect the EEDI score. If combining PTO/ PTI, the normal operating mode should be used in EEDI. This is usually the PTO mode, as it is the most efficient mode and it more closely resembles the normal DNV GL Report No , Rev. 0 Page 3 H:\MEPC\72\MEPC 72-INF.12-annex 10

220 Annex 10, page 8 sea going condition. Under development. 11 Fuel cell propulsion 12 Selection of coatings 0 % 0 % 0 % 0 % 2 % 2 % 2 % 2 % 13 Air lubrication 1 % 1 % 0 % 0 % 14 LNG % % % % If using fuel cells for propulsion the CO 2 emissions would be close to zero, if the production of hydrogen is disregarded. Large potential for ships in operation, but limited effect on the EEDI score as most hulls are smooth and without fouling during the sea trial. Larger potential for resistance reductions for vessels with higher speeds, but this also requires more air, i.e. a larger system energy use. A combination of air lubrication and coating is estimated to give a 2 % reduction potential for all the ship types. The availability of LNG infrastructure and supply could be problematic. DNV GL Report No , Rev. 0 Page 4 H:\MEPC\72\MEPC 72-INF.12-annex 10

221 Annex 10, page 9 The feasibility of the measure depends on trade area. Using LNG as fuel and its effect on the EEDI score is modelled by DNV GL assuming SFCs [g/kwh]: SFCME=140 SFC pilot fuel ME=6 SFCAE=180 SFC pilot fuel AE=7 The availability of fuel can be problematic. 15 Biofuels Not quantified Not quantified Not quantified Not quantified 16 LPG Not quantified Not quantified Not quantified Not quantified 17 Alcohol Not quantified Not quantified Not quantified Not quantified Quantificatio n of the potential for EEDI score improvement is not evaluated for this fuel since the potential for LNG is considered larger. Quantificatio n of the potential for EEDI score improvement is not evaluated for this fuel since the potential for LNG is considered larger. The availability of ethanol/meth anol supply could be problematic. Quantificatio n of the potential for EEDI score improvement is not evaluated for DNV GL Report No , Rev. 0 Page 5 H:\MEPC\72\MEPC 72-INF.12-annex 10

222 Annex 10, page 10 this fuel since the potential for LNG is considered larger. Problematic for ships with little available deck space. 18 Wind power Uncertain effect on the EEDI calculation. Omitted in the calculation of effeciency improvement potential. Uncertain effect on the EEDI calculation. Omitted in the calculation of effeciency improvement potential. Uncertain effect on the EEDI calculation. Omitted in the calculation of effeciency improvement potential. Uncertain effect on the EEDI calculation. Omitted in the calculation of effeciency improvement potential. Evaluating the effect of wind power on the EEDI will require further analysis, which is of such an extent that it would be outside the scope of this project. 19 Solar power 1 % 1 % 0 % 0 % 20 Hull vane 0 % 1 % 3 % 3 % Total reduction potential Without LNG: 12 % With LNG: approx. 32 % Without LNG: 12 % With LNG: approx. 33 % Without LNG: 13 % With LNG: approx. 33 % Without LNG: 14 % With LNG: approx. 34 % Problematic for ships with little available deck space, thus deemed not relevant for the two smallest ship sizes. Limited effect on slow and large vessels. DNV GL Report No , Rev. 0 Page 6 H:\MEPC\72\MEPC 72-INF.12-annex 10

223 Annex 10, page 11 3 EVALUATION OF MINIMUM PROPULSION POWER AND THE SHIP EFFICIENCY IMPROVEMENT POTENTIAL FOR TANKERS AND BULK CARRIERS To evaluate the potential for efficiency improvements the minimum installed propulsion power for tankers and bulk carriers must also be considered. The minimum propulsion power requirement is meant to assure that tankers and bulk carriers are able to maintain manoeuvrability in adverse conditions. There is more than one way to check this requirement, but the crudest and most conservative way of evaluating it is using the level 1 assessment of minimum power lines. For different ship types the minimum power line values of total installed MCR [kw] is found through Minimum power line value = a (DWT) + b where the a and b parameters are found in Table 3-1. Table 3-1 Parameters of the minimum power lines Ship type a b Bulk carrier which DWT is less than 145, Bulk carrier which DWT is 145,000 and over Tanker Combination tanker see tanker above For the reference vessels selected for both tanker categories and the Panamax bulk carrier, all the total installed propulsion powers are below the minimum power line. The installed propulsion power to minimum propulsion power ranged between approximately 80 % to 95 % for the different vessels. As the case studies in this project are based on reference ships that have the lowest EEDI scores of DNV GL classed vessels within each category, it was expected that the installed propulsion power for each vessel was as low as possible in order to achieve the best possible EEDI score. Assessing the minimum propulsion power with a less conservative method, e.g. level 2, is too comprehensive for the project scope. However, DNV GL believe that the installed propulsion power is very close to the currently calculated minimum propulsion power for each of the examined vessels. This means that in order to apply further energy saving devices to these vessels they either have to apply additional to reduce the minimum propulsion power (e.g. CPP propeller or high lift rudder) or take out the potential in increased EEDI trial speed. The speed increase can be calculated by setting: V = V ref 3 1 reduction potential from measures The estimated total reduction potentials for each vessel are presented in Table 2-1. V ref is the EEDI trial speed for the reference vessels. Using the reduction potentials (excluding LNG as fuel) for the tankers and the bulk carrier, the speed increase is found to be 4.4 % for the VLCC and Panamax bulk carrier and 4.8 % for the chemical/product tanker. DNV GL Report No , Rev. 0 Page 7 H:\MEPC\72\MEPC 72-INF.12-annex 10

224 Annex 10, page 12 4 GENERAL CARGO SHIP EFFICIENCY IMPROVEMENT POTENTIAL The general cargo ship does not have EEDI requirements for minimum installed propulsion power. Therefore, the efficiency improvement potential can be evaluated at the same EEDI speed. As presented in Table 2-1, the estimated potential is 14 % when alternative fuels are not considered and approximately 34 % in total when including LNG as fuel. 5 SUMMARY OF EEDI SCORE POTENTIALS The case study results are presented in Table 5-1. The general cargo vessels already comply with phase 3 requirements, the Panamax bulk carrier and the chemical/product tanker are close to satisfying phase 3, while the VLCC requires significant improvements to meet phase 3 requirements. Applying the identified, LNG and taking out the benefit as increased EEDI speed for the bulk carrier and tankers, the EEDI score reduction relative to phase 0 can be improved by more than % for all reference vessels. Excluding LNG as an option the potential is more than % for all vessel types relative to phase 0. The general cargo ship seems to have an especially large potential for efficiency improvements, up to % when utilizing both energy saving devices and LNG as fuel. Table 5-1 EEDI score reduction relative to phase 0 requirement Panamax bulk Chemical/product VLCC tanker carrier tanker General cargo ship Basecase % % % % With reduction measures*/increased speed** (excl. LNG as fuel) With LNG as fuel only (main and auxiliary engines) With reduction measures/increased speed (incl. LNG as fuel (main and auxiliary engines)) % % % % % % % % % % % % * Applied for the general cargo ship ** Applied for the two tankers and the bulk carrier DNV GL Report No , Rev. 0 Page 8 H:\MEPC\72\MEPC 72-INF.12-annex 10

225 Annex 10, page 13 6 CONSIDERATIONS ON REQUEST 2 The second part of this project is to briefly comment on the Requests 2.1 to Request 2.4. Not studies have been conducted in this part, only a workshop to gather initial opinions. Request 2-1 There have been opinions that phase 3 could be difficult to achieve for tank and bulk vessels. This concern is believed to be caused by the need to satisfy both the EEDI and the minimum propulsion power requirement. For the minimum propulsion power requirements, a provision was added to regulation 21 in chapter 4 of MARPOL Annex VI, stating that: For each ship to which this regulation applies, the installed propulsion power shall not be less than the propulsion power needed to maintain the manoeuvrability of the ship under adverse conditions [ ]. For bulk carriers and tankers a guideline on how to compute the minimum propulsion power is developed [REF]. This result in a more stringent interpretation of the minimum propulsion power for these vessels types than for others. The easiest way of improving EEDI for most vessels is to reduce the installed power, however current designs are close to the minimum propulsion power requirement and are not satisfying the EEDI phase 3 requirements. This means that complying with EEDI is not possible by simply reducing the installed power. Assuming a recent, but traditional design is close to the minimum propulsion power limit there are three main options for improving the EEDI: 1. Improving hull and propulsion design to achieve higher speed using the same engine power 2. Alternative fuels 3. Improve performance in harsh weather to reduce minimum propulsion power requirement and hence reduce installed power It should be noted that efficiency improvement taken out as speed increase only has about 1/3 effect on the EEDI as if the gain was taken out as reduced installed power. The following options, related to hull and propulsion design (i.e. the first point (1) in the list above), can be used to improve EEDI: Alternative fuels (can have a large effect on the EEDI) Reduce installed power Hull and propeller optimization Improved machinery efficiency through design or operation optimization Waste heat recovery PTO/PTI - If you use a combined PTO/ PTI, the normal operating mode shall be used in EEDI. This means you can install a large PTI for use in bad weather or to catch up delays without being penalized in EEDI. When PTO/PTI is installed, normally the PTO mode is applied in EEDI DNV GL Report No , Rev. 0 Page 9 H:\MEPC\72\MEPC 72-INF.12-annex 10

226 Annex 10, page 14 calculations because it is the most efficient mode and it more closely resembles normal sea going condition, which is what EEDI is supposed to be used for. Air bubble lubrication does not affect the limit for minimum propulsion. However, it will have a positive effect on the EEDI on sea trial. Possibility for further technology development If minimum propulsion power/manoeuvrability is an issue the examples below can allow bulk carriers and tankers to reduce the required minimum propulsion power and hence meet the EEDI requirements by reducing the installed power. The following list of measures thus complements the third point (3) in the list of options for reducing the EEDI on the previous page: CPP will increase thrust at low speeds (able to take out full power) - FPPs as opposed to CPPs can usually not take out full power at low speeds due to torque limitations. - CPPs can produce up to 50% more thrust in bollard pull than FPPs. Ducts will increase produced thrust at low speeds, but reduce thrust at high speeds High lift rudders e.g. flap rudders, fish tail rudders or similar will increase steering force - High lift rudders e.g. flap rudders generate up to 50% more steering force than conventional rudders. Azimuth propulsion can produce full thrust in any direction. Reduce added resistance in a seaway. In addition, ice correction factors can be relevant for bulk carriers and tankers if built within the Common Structural Rules (CSR) regulations. To summarize: Without a detailed study, it is believed that the phase 3 requirements can be met without introducing a correction factor for tankers and bulk carriers. In addition to alternate fuels, improving the performance in harsh weather and hence reducing the minimum propulsion power requirement is a promising option. Request 2-2 By building a vessel with LNG as fuel the EEDI can be reduced by approximately 25 %. Today 118 vessels run on LNG, 121 projects are on order. There are 63 ports where LNG can be bunkered worldwide with 26 decided to be build (ref. LNGi DNV GL). To assess whether all smaller container ships can run on LNG a feasibility study must be conducted world-wide to assess the cost of installing the infrastructure and which barriers need to be overcome to build the LNG infrastructure. By lowering the speed, smaller container ships that do not meet the phase 3 requirements today will be able to meet the EEDI requirement. However, building small container vessels that sail with lower speeds might affect the competition against trailers, especially for perishable cargo that requires quick transport DNV GL Report No , Rev. 0 Page 10 H:\MEPC\72\MEPC 72-INF.12-annex 10

227 Annex 10, page 15 and delivery. A shift from sea to road transport would lead to higher carbon dioxide emissions, not benefiting the climate or the environment. See also Request 2-3. Request 2-3 It has been considered that reducing the speed is the best measure to improve ship efficiency. It is included in the current formulas for EEDI with the engine load at 75 MCR. When considering the reduction of ships design speeds, it is important to evaluate the overall environmental accounting. Some reflections that might question very low sailing speed for ships: By reducing the sailing speed of the smaller vessels, trucks could, for more trade routes, become a quicker way to transport cargo. Trucks are a less effective means of transport compared to ships when disregarding the very short distances. With a slower-sailing fleet more vessels might be required to meet the need for transport which s would affect the environmental accounting. Old vessels will be able to compete with new vessels as they are not restricted to EEDI. The initiatives of EEDI and EEOI are positive when it comes to emissions, but taxation of emissions would lead to even larger reductions. However, intercontinental sea transport has a long way to go in order to achieve a similar arrangement. Request 2-4 Today, batteries, and thus hybridization, are not included in EEDI. Batteries can reduce energy use and emissions, e.g. through peak shaving or when manoeuvring. When calculating EEDI, only the energy carrier that has the highest burning value is included in the calculation. Batteries in a hybrid solution is therefore not included. This is a field that should be investigated further. Some information about hybridization: Batteries are still not treated in EEDI context The batteries can be charged and the stored energy can be used when the ship needs more power in e.g. adverse conditions Batteries can be charged either via PTOs on ME, WHRs on ME/AE, brakes for crane lowering, etc. Container vessels can achieve large efficiency improvements with peak shaving for refrigerated containers. Power can be used for e.g.: - Adverse conditions - Cargo handling - Extra power for higher speed - peak loads which normally need larger AEs to handle, or more AEs in operation DNV GL Report No , Rev. 0 Page 11 H:\MEPC\72\MEPC 72-INF.12-annex 10

228 Annex 10, page 16 - Fuel savings through running AE on/off and charging the batteries When PTO/PTI is installed, normally the PTO mode is used in EEDI calculations, because it is the most efficient mode and corresponds to the normal sea going condition, which EEDI is supposed to be used for. If you use a combined PTO/ PTI, the normal operating mode should be used in EEDI. This means you can install a large PTI for use in bad weather or to catch up delays without being penalized in EEDI. In today s formula, power take off (PTO) power take in (PTI) has a negative effect on the score. If using a PTI it can count it as a PTO if it is used as a PTO. A PTI/PTO solution can be used to reload speed. Boost at delay may be an additional effect when calculating the EEDI. DNV GL Report No , Rev. 0 Page 12 H:\MEPC\72\MEPC 72-INF.12-annex 10

229 Annex 10, page 17 7 REFERENCES /1/ Annex 4 Summary of latest information of energy saving DNV GL Report No , Rev. 0 Page 13 H:\MEPC\72\MEPC 72-INF.12-annex 10

230 H:\MEPC\72\MEPC 72-INF.12-annex 10 MEPC 72/INF.12 Annex 10, page 18

231 Annex 10, page 19 About DNV GL Driven by our purpose of safeguarding life, property and the environment, DNV GL enables organizations to advance the safety and sustainability of their business. We provide classification and technical assurance along with software and independent expert advisory services to the maritime, oil & gas and energy industries. We also provide certification services to customers across a wide range of industries. Operating in more than 100 countries, our professionals are dedicated to helping our customers make the world safer, smarter and greener. H:\MEPC\72\MEPC 72-INF.12-annex 10

232

233 Annex 11, page 1 Analysis of 2017 EEDI Database Source: EEDI Review CG Round 2; Annex 5 EEDI_database_Report_ xlsx Provided by the Secretariat Database as of 31 October 2017 (2,671 ships) H:\MEPC\72\MEPC 72-INF.12-annex 11

234 Annex 11, page 2 Comparison to previous version 3/31/17 (MEPC 71/INF.14) 10/31/17 Total ships 2,443 2,671 Bulk Carrier 1,181 1,266 Gas carrier Tanker Containership General Cargo Refrigerated Cargo 2 2 Combination carrier - - LNG carrier - - Ro-Ro Cargo (vehicle carrier) Ro-Ro Cargo 9 13 Ro-Ro passenger - - H:\MEPC\72\MEPC 72-INF.12-annex 11 Cruise - -

235 Annex 11, page 3 General Observations New data does not change observations from the previous data set There continues to be many ship designs that do not achieve the Phase 3 30% reduction The range of results within ship types varies by ship size New data on innovative suggests uptake is slow Of 228 additional ships, only 29 use innovative 24 of these are container ships 14: waste heat recovery 10: other H:\MEPC\72\MEPC 72-INF.12-annex 11

236 Annex 11, page 4 General Observations There is a large amount of data for 4 ship types But the increase in number of ships compared to the 2016 database is not large Results of 4 main ship types Bulk carriers Average EEDI is only slightly better compared to 2016 database, but spread is also slightly less Only bulk carriers 40,000 to 74,900 DWT meet the 20% reduction, on average Very few bulk carriers meet the 30% reduction Tankers Average EEDI is only slightly better compared to 2016 database, but spread is also slightly less Tankers <225,000 DWT meet the 20% reduction, on average Tankers >225,000 DWT do not meet the 20% reduction Some tankers meet the 30% reduction, but most do not Containerships Average EEDI is better compared to 2016 database, but spread is larger Most containership designs meet the 20% reduction target, although there are still some designs that don t Many containerships meet the 30% reduction, but a significant number do not Gas Carriers Average EEDI is better compared to 2016 database, but spread is larger Very few gas carriers meet the 30% reduction H:\MEPC\72\MEPC 72-INF.12-annex 11

237 Annex 11, page 5 General Observations There is only a small amount of data for 4 ship types General Cargo The database is still very small: 37 ships (there are 15,776 ships in Sea-Web) Average EEDI is better compared to 2016 database, but spread is larger Refrigerated cargo: 2 ships One is 39.2% better than the reference line, the other is 12.6% better RoRo Vehicle: 23 ships, none of which are subject to reduction requirements RoRo Cargo: 9 ships, none of which are subject to reduction requirements There is no data at all for 4 ship types: Combination carrier LNG carrier RoRo Passenger Cruise ship w/nonconventional propulsion H:\MEPC\72\MEPC 72-INF.12-annex 11

238 Annex 11, page 6 Bulk Carriers Bin values are the upper bounds SeaWeb 2014: 10,755 EEDI: >10,000 H:\MEPC\72\MEPC DWT 72-INF.12-annex 11 Number of Ships Average St. Deviation 2015 data set % 6.0% 2016 data set % 5.1% 3/31/17 data set % 4.8% 10/31/17 data set % 5.0%

239 Annex 11, page 7 Bulk Carriers by sub-category Number of Ships Average Standard Deviation March Oct March Oct March Oct 10,000 39,900 DWT % 18.3% 6.1% 6.1% 40,000 74,900 DWT % 20.9% 4.1% 4.1% 75, ,900 DWT % 18.9% 4.8% 5.6% >175,000 DWT % 17.6% 3.8% 3.8% EEDI applies ships >10,000 DWT Only ships 40,000-74,900 DWT meet 20% on average There may be issues for the other sub-category designs H:\MEPC\72\MEPC 72-INF.12-annex 11

240 Annex 11, page 8 Tanker Bin values are the upper bounds SeaWeb 2014: 12,462 (oil and chemical) EEDI: >4,000 DWT H:\MEPC\72\MEPC 72-INF.12-annex 11 Number of Ships Average St. Deviation 2015 data set % 8.3% 2016 data set % 8.2% 3/31/17 data set % 8.1% 10/31/17 data set % 8.1%

241 Annex 11, page 9 Tankers by sub-category Number of Ships Average Standard Deviation March Oct March Oct March Oct 4,000-59,900 DWT % 26.7% 7.2% 7.5% 60, ,000 DWT % 21.5% 7.2% 6.8% >225,000 DWT % 14.9% 4.7% 4.7% EEDI applies ships >10,000 DWT Only ships 40,000-74,900 DWT meet 20% on average There may be issues for the other sub-category designs H:\MEPC\72\MEPC 72-INF.12-annex 11

242 Annex 11, page 10 Containership Bin values are the upper bounds SeaWeb 2014: 5,040 EEDI: >10,000 DWT H:\MEPC\72\MEPC 72-INF.12-annex 11 Number of Ships Average St. Deviation 2015 data set % 11.6% 2016 data set % 11.5% 3/31/17 data set % 12.5% 10/31/17 data set % 12.3%

243 Annex 11, page 11 Containerships by sub-category Number of Ships Average Standard Deviation March Oct March Oct March Oct 10,000-49,500 DWT % 29.7% 11.6% 11.4% 50,000-99,500 DWT % 35.7% 13.3% 12.6% 100, ,500 DWT % 41.5% 10.5% 10.4% >180,000 DWT % 48.0% 5.4% 5.2% EEDI applies ships >10,000 DWT Ships generally meet the 30% Phase 3 required reduction, on average, but some ships designs still do not H:\MEPC\72\MEPC 72-INF.12-annex 11

244 Annex 11, page 12 Gas Carrier Bin values are the upper bounds SeaWeb 2014: 1,683 EEDI: >2,000 DWT H:\MEPC\72\MEPC 72-INF.12-annex 11 Number of Ships Average St. Deviation 2015 data set % 8.2% 2016 data set % 8.8% 3/31/17 data set % 9.5% 10/31/17 data set % 9.8%

245 Annex 11, page 13 General Cargo Bin values are the upper bounds SeaWeb 2014: 15,776 EEDI: >3,000 DWT H:\MEPC\72\MEPC 72-INF.12-annex 11 Number of ships Average St. Deviation 2015 data set % 20.1% 2016 data set % 19.6% 3/31/17 data set % 20.0% 10/31/17 data set % 19.9%

246

247 Annex 12, page 1 DWT Number of NB Average EEDI Energy saving arrangements Operational difficulties accelleration problems in bad weather, low Light Runnining ,74 pre swirl fins, low friction anti fouling, rudder bulp Margin of the propeller, propeller edge cutting, Vessel lost speed in Heavy weather due to limited power reseve ,00 pre swirl fins, rudder bulp ,87 Duct and rudder bulp Engine slow down in heavy weather due to overload ,83 Engine tuning PL EGB with EEC and aux engine boiler accelleration problems in bad weather, low Light Runnining Margin of the propeller, propeller edge cutting ,57 Mewis duct, low friction anti fouling accelleration problems in bad weather, low Light Runnining Margin of the propeller, propeller edge cutting ,20 Mewis duct, low friction anti fouling, A/E boiler ,18 Lines development, propeller development, Mewis duct,rudder bulp, engine tuning, low friction anti fouling, A/E boiler low Light Runnining Margin of the propeller 14,00 EEDI Requirements Bulker 12,00 10,00 8,00 6,00 EEDI [gco2/ t sm] 4,00 2,00 0, DWT [t] Reference Oldendorff NB series Phase 1 Phase 2 Phase 3 H:\MEPC\72\MEPC 72-INF.12-annex 12

248

249 MEPC 71/INF.12 Annex 13, Page 1 ANNEX 13 CONTRIBUTION TO REQUEST 4 OF THE SECOND ROUND BY THE RUSSIAN FEDERATION Categorization / Type of 1 Design improvements Short-description ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. (1) Optimization of hull (Revised by CESA) There are many barriers to focusing solely on modifying the hull lines to achieve morefavourable resistance. Examples are the effects of the given requirements of the amount and type of payload and the dimensions of ports and terminals. These barriers will considerably reduce the potential for the reduction of resistance and of fuel consumption. On single ships, improvements in power requirements of up to 30% have, in fact, occasionally been achieved on particularly illconceived designs; however, the mean potential for improvement would be expected to be small. However, some potential exists in the optimization onto operational profiles instead of a single design condition. Smaller ships are more sensitive to design details, since they have comparatively large wave-generating resistance and also because less resources will traditionally be available for optimization, due to the smaller overall budget that is typically available for developing the designs. The expected improve ratio depends on an individual ship. Resistance and energy consumption increase during ship operation in a natural seaway. Traditionally, ships have been optimized primarily for calm-water conditions in a towing tank (not least because the contractual measurements of trial performance are conducted in calm water); however, optimization for irregular wave conditions is becoming more common. During their lifetime, ships will more frequently operate in a wave field that is characterized by the short wavelength λ (small sea (Japan) 0-3% (Japan) This technology has been improved for a long time and there are few room remained for further improvement. (WSC) Please note that optimization of the hull is reflected in the Vref (speed) at scantling draft on the chosen power. This draft is however (CESA) can be combined with most of the other without loss of improvement ratio, in particular categories 1, 2, 4, 5,6(1)(2) and (4) and 7(2). The improvement potential of of category 3, 6(3) and 7(1) might be slightly reduced. For ice class 1ASuper and equivalent (PC6) optimization of hull is possible but limited by ice class rules. For ice class higher than 1ASuper (starting from Arc 5 in Russian Register, corresponded to PC6) application of icebreaking bow and limitation on stern hull lines (prohibited transom) are obligatory and lead to increasing of resistance at full speed up to 20-30%. Optimization is performed aiming mainly ice resistance and ice capability, not open water resistance. Measure designed for relatively high speed slender ship is not a subject of discussion because design of ice class ships is in contradiction with design of H:\MEPC\72\MEPC 72-INF.12.docx

250 MEPC 71/INF.12 Annex 13, Page 2 Categorization / Type of Short-description states) in comparison to the ship length L. Therefore, optimization for waves generally emphasizes short-wavelength waves (this however holds only for large vessels, smaller ships (feeders, coastal ) do operate in conditions λ/l 1). There is an example in relation to the technology for the optimization of hull, namely, Stern Wedge. Example 1 Stern wedge By guiding the propeller flow a bit down a lift is created in the aft part of the hull which can reduce the engine power. The optimum slope of the wedge is often appr. 10 degrees. This type of improvement is normally only beneficial for slender and relatively fast ships (typical navy ships and fast patrol vessel or similar) A power reduction of up to 10 % can be obtained. ratio (%) for may ship types rarely utilized. If optimizing the hull lines for this draft to achieve a better EEDI, the performance on actual drafts may not necessarily be optimum. ratio when using with other energy saving (%) Expected improvements from holistic design optimisations (hydrodynamic hullform only) are expected to achieve full 5% - 7% improvements for the entire operational envelop. Perspective of application for ice class ships. fast speed ships (for example the needs in powerful bossings for ice class ships inevitable caused propeller cavitation problems). There are some applicable for certain types of ships. Superyachts are generally characterized by Loa between 40 and 100 meter, high Froude numbers (>0.35) and slender lines (CB<0.5). Most yachts are full displacement vessels; a limited number is (semi)planning. Motion behaviour is an important design criterion in connection with required levels of comfort. Therefore hullform innovations are generally not only aiming at reducing resistance but also improving sea keeping properties. However, recent hullform innovations result also in (considerable) reductions of hull resistance, not only in adverse sea conditions, but also in still-water conditions. Recent hullform innovations for superyachts comprise: Axe bow ( J.L. Gelling, "The Axe Bow", HISWA Symposium 2006) Supersport hullform ( ) Fast Displacement Hull Form ( H:\MEPC\72\MEPC 72-INF.12.docx

251 MEPC 71/INF.12 Annex 13, Page 3 Categorization / Type of Short-description displacement-hull/ and Perry van Oossanen et al, "Motor Yacht Hull Form Design for the Displacement to Semi-Displacement Speed Range", FAST 2009 Conference) All above hull forms are applied on newbuilds of SYBAss members and are currently in service. For confidentiality reasons we have provided publicly available information only (refer to the above websites and reference papers). The are also applicable for other ship types of comparable characteristics and/or type of operation, e.g.: the axe bow has also been applied on a number of fast crew suppliers, windfarm maintenance vessels, fast patrol and Search & Rescue vessels. ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. (2) Optimization of superstructure Optimization of the superstructure of ships for reduced resistance to still air and to wind has traditionally not been an important subject. Again, there are barriers, in the form of requirements for and the usage of covered spaces. However, for ships with large superstructures and for ships operating at relatively high speeds, there will be a potential for reduction of power consumption by carrying out systematic streamlining of the superstructure to the greatest possible extent. For these ships, it is estimated that there is potential for reduction in power consumption of 2-5%, depending on the size of the superstructure and the area in which the ship operates. Also, for other ships, there is expected to be a certain potential for reduction in power consumption, perhaps in the order of 1-2%, by keeping the topsides as uncluttered and streamlined as possible. The efforts to achieve reductions may range from the simple (such as grinding weld beads flat) to the more extensive (for example, redesigning and repositioning cranes, applying spoilers to alter the airflow over the funnel and deck-houses, and designing more streamlined deck-houses).in the case of transition to electric population (Japan) Approximately 0-2% reduction of power consumption, which corresponds to about 5-30% reduction of wind resistance. For some type of ships, 3% reduction of the power consumption has been achieved. (CESA) can be combined with all without loss of improvement ratio with the exception of some of the wind power options 6(3) Applicable for ice class ships. H:\MEPC\72\MEPC 72-INF.12.docx

252 MEPC 71/INF.12 Annex 13, Page 4 Categorization / Type of Short-description system which will be able to transfer engine room to bow-side, streamlining superstructure will be adopted. ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. (3) Optimization of propeller (Japan) Integrated streamline hull and superstructure form for PCC PCC has very tall and large ship s main body above water surface from fore-end to aft because of the purpose of this ship type. This is why wind resistance is large, and fuel consumption due to wind resistance becomes very large particularly under rough sea conditions. Integrated streamline hull and superstructure form could reduce wind resistance when receiving head wind and could assist fore thrust when receiving bow wind, and this could reduce fuel consumption. (Revised by CESA) The main abating effect of optimization of the propeller is obtained by increasing the diameter of the propeller and reducing the number of its revolutions per minute (RPM).The requirements to maintain adequate clearances between the propeller and the hull and to attain sufficient submersion of the propeller when the ship is operating in a seaway and/or in ballast condition set restrictions on the extent to which the diameter of a propeller can be increased. A propeller that is operating at a low number of revolutions per minute may require the additional cost of installing a reduction gear, while propellers operating at a higher number of revolutions per minute can generally be directly connected to the main engine. Propellers with a large diameter, operating at a low number of revolutions per minute, will therefore be best suited to deepdraught ships; this includes most tankers and bulk carriers and many general cargo vessels. Such propellers will be less suited to many container vessels, and they will not, in general, be suited to RoRo cargo, RoPax vessels or cruise vessels. (Japan) 0-3% (The Contracted Loaded Tip (CLT) propeller shows the higher improvement ratio, which depends on the ship type, the propeller loading coefficient etc.) (Japan) (Japan) This could interfere with other energy saving devices especially for devices fitted on stern part of ship. (CESA) can be combined with most of the other without loss of improvement ratio, in Propellers for high ice class ships as the first priority should provide reliability in ice and usually have more thick blades, limitation on the,measure against cavitation. Main design mode for high ice class propeller is bollard pull to provide maximum bollard pull. For medium and light ice classes (below 1ASuper and equivalent) propeller optimization for open water condition are applicable but limited by ice class requirements comparing with propellers without ice class. H:\MEPC\72\MEPC 72-INF.12.docx

253 MEPC 71/INF.12 Annex 13, Page 5 Categorization / Type of Short-description Existing ships of the displacement type mostly operate within a moderate Froude number interval (Fn 0.3). This fact dictates the choice of propulsive devices (propellers) that may be used in order to achieve ship thrust. The figure below shows the efficiency as function of the propeller loading coefficient CTh, which is from "Hydrodynamics of Ship Propellers" by Andersen and Breslin (Cambridge University Press 1994). The thrust loading coefficient CTh is defined as follows: T C Th = 1 and C 2 ρ A 2 Th = 8 R disk V A π (1 t) ρ (V A D prop ) 2 C Th = 8 π K T J 2 A disk = π 4 D 2 propj = V A n D R K T = (1 t) ρ n 2 R = (1 t) T V D4 A = (1 w) V prop ratio (%) Ratio is updated based on the actual performance measured. ratio when using with other energy saving (%) particular categories 1, 2, 4, 5 and 6. The improvement potential of category 7(1) might be slightly reduced. Combination with category 3 is not applicable. Perspective of application for ice class ships. Measures against cavitation (skew, tip unloading, blade profiles) for ice class propeller (especially for high ice class) are strongly limited by the requirements to propeller ice operation. New type of propellers such as CLT are not applicable even for light ice class ships because of danger that blades will be bent in operation even in light ice (similar to sour experience of numerous cases of skewed blades bending during light ice operation). where: Dprop = propeller diameter V = ship speed J = propeller advance ratio R = ship resistance T = propeller thrust t = thrust deduction fraction w = wake fraction = density of sea water H:\MEPC\72\MEPC 72-INF.12.docx

254 MEPC 71/INF.12 Annex 13, Page 6 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. Figure Efficiency of propulsive devices A weaker but nevertheless considerable lever to improve propeller efficiency is the concentration of its radial load distribution close to the blade tips. This measure is limited by the concurrent increase of vibration excitation but is frequently used in case of retrofit propellers in times of slow steaming. H:\MEPC\72\MEPC 72-INF.12.docx

255 MEPC 71/INF.12 Annex 13, Page 7 Categorization / Type of Short-description The most efficient type of propeller is a well-designed fixed-pitch (helical)propeller. However, for other reasons, alternative propulsion devices need to be considered. For instance, controllable-pitch (CP) propellers, although slightly less efficient then fixed-pitched propellers, may be selected if the ship in question needs to satisfy the requirements of rapidly reversing thrust or efficient operation in significantly different environmental conditions. On the other hand, for ships with demands for high manoeuvrability, propellers with a vertical axis or CP propellers represent a preferable choice. ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. There are two examples in relation to the technology for the optimization of propeller, namely, Kappel propeller and CLT propeller. Example 1 Kappel propeller Tip vortices are formed due to the difference in pressure between the pressure and suction side of the propeller as the water will move from the region of high pressure to the region of low pressure. The pressure on both sides near the tip will therefore equalize and the efficiency of the tip region will decrease. The Kappel propeller minimizes the flow over the tip, and the outer region of the Kappel propeller therefore retains a high efficiency increasing 4-5% of the total efficiency of the Kappel propeller compared to conventional propellers. eller Example 2 CLT (Contacted Loaded Tip) propeller CLT propellers are characterized by the following: The blade tip generates a substantial thrust. The pitch increases from the root to the tip of the blades. The chord at the tip is finite. H:\MEPC\72\MEPC 72-INF.12.docx

256 MEPC 71/INF.12 Annex 13, Page 8 Categorization / Type of Short-description End plates are fitted at the blade tips, toward pressure side; they are adapted to the fluid vein contraction to reduce as much as possible their viscous resistance. ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. The end plates operate as a barrier, avoiding the communication of water between the pressure and the suction side of the blades, allowing to establish a finite load at the tip of the blades. The fundamental goal of the CLT propeller is to improve the propeller open water efficiency by reducing the hydrodynamic pitch angle through the reduction of the magnitudes of induced velocities at the propeller disk. The pressure drop on the suction side of CLT propellers is less than for the conventional equivalent and therefore the extent of the cavitation developed on the suction side is lower and hence the pressure forces that a CLT propeller exerts on the stern hull structure are lower. Additionally, CLT propellers have a reduced tip vortex because of the existence of the end plates. The combination of these circumstances means that the pressure forces exerted by a CLT propeller on the stern structure are of a lower magnitude than for conventional propellers, and so in turn the induced hull vibration and noise levels on board are lower.the noise radiated to the water is also lower for CLT propellers. As the blade area of the CLT propeller is more efficiently used for supplying thrust, the optimum diameter is lower than for an equivalent conventional propeller. CLT propellers offer higher efficiency which may be used to achieve fuel savings at constant ship speed or alternatively ship speed increase at constant fuel consumption. H:\MEPC\72\MEPC 72-INF.12.docx

257 MEPC 71/INF.12 Annex 13, Page 9 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. (4) Podded (azipod) drives Podded (azipod) drives are systems where an electric motor with a propeller is suspended under the appropriate (usually the aft) section of the hull. The pod can be rotated to direct the thrust, resulting in very good ship manoeuvrability. In many cases, more than one pod is used. New pod drives have pulling propellers that face forward. This gives the pod a good flow of water into the pod, resulting in high propulsion efficiency. However, the pod in itself increases the drag, thus reducing total efficiency. Experience from tests of hulls in the towing tank at MARINTEK clearly indicates that the net effect of podded propulsion on the energy efficiency of propulsion is generally negative when compared to conventional designs of propulsion systems. Similar to the effect of conventional propeller. (CESA) can be combined with most of the other without loss of improvement ratio, in particular categories 1, 2, 4, 5 and 6. The improvement potential of Podded propulsors now are installed to the ships even of high ice class. For high ice class large transportation ship podded propulsors are the best and may be only way to provide manoeuvrability in ice. But open water propulsion efficiency of podded propulsor is similar to conventional propellers and podded propulsors is not considered as energy saving measure. H:\MEPC\72\MEPC 72-INF.12.docx

258 MEPC 71/INF.12 Annex 13, Page 10 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) category 7(1) might be slightly reduced. Perspective of application for ice class ships. (Netherlands) (5) Matching and integration of hull form and drivetrain An integral approach, system integration, of hullform and drivetrain (propeller, gearbox, main engine) could lead to higher optimisations of solely improvement on a single system. The complete performance of the hull-propeller interaction, driven by the engines, could be assessed in a way where all expected operational conditions are considered. When the whole system is seen as systems which interact and interfere with each other, the optimal solution can be found. Experience from Conoship International shows that proper integration of a hull form (with the ConoDuctTail), nozzle, propeller, gearbox and main engine leads to savings of abt. 20%, with even better results in seaways. 1-20% For medium and high ice class ships propulsion complex is design mainly to prevent overloading and blockage in ice. Ships of high ice class always design with electric power drive with lower drive efficiency than for straight drive from diesel to propulsor. So, system integration usually take place, but with another target than open water efficiency. 2 Recovery of propeller energy A considerable number of devices have been invented for improving the power consumption of ships by recovering as much as possible of this rotational energy in the flow from the propeller, or to provide some pre-rotation of the inflow into the propeller. The most important of these will be considered here. H:\MEPC\72\MEPC 72-INF.12.docx

259 MEPC 71/INF.12 Annex 13, Page 11 (1) Coaxial contra-rotating propeller The coaxial contra-rotating propeller is an obvious device for recovering some of the rotational energy. To avoid problems with cavitation, the aft propeller usually has a smaller diameter than the front propeller. The aft propeller is therefore not working on the complete rotating flow field from the forward propeller. In addition, the more complicated shafting results in mechanical losses that offset some of the gain that is obtained by recovering the rotational energy. It is also reported that gearboxes for contra-rotating propellers may present problems. Reported gains in power consumption range from 6% to 20%. Gains of 15% and 16% have been reported from two different full-scale measurements. The improve ratio is affected by the propeller loading (thrust per unit area), different ratios of the diameters of the two propellers, or some other factor(s), and cannot be achieved for every ship. Contra-rotating propeller arrangements require a short shaft line and are therefore primarily suited to single-screw ships. The arrangement is particularly beneficial for relatively heavily loaded propellers, and the best results (in the form of power consumption) have been found in fast cargo vessels, ro ro vessels and container vessels. Naturally, this type of technology is tested in cases where it is expected to be particularly suitable. The contra rotating propellers have also been used for podded propulsors, where the same efficiency gain of 5 10 % has been obtained. 8% (CESA) (Japan) This technology collects the energy of rotational flow. This is why this could interfere with other energy saving devices which have same effect such as pre-swirl devices and post-swirl devices. (CESA) can be combined with most of the other without loss of improvement ratio, in particular categories 1(1)(2), 4, 5 and 6(1)(2)(3). The improvement potential of combinations within category 2 and with 7(3) might be Maximum gain to efficiency due to CRP corresponds the case of fast speed ships when single propeller (as a rule with limitation of diameter) operate in thrust break cavitation mode in this case, distribution of propeller loading between two CRP is favourable. For lower speed ships gain of efficiency is lower than 10% comparing with high quality single propeller design. Couple Pod + propeller on shaft demonstrates lower efficiency by 3-4% comparing with CRP on the shaft due to resistance of pod. In any case shaft line for conventional CRP (both propeller on the shaft) have extremely complicated shaft line. Taking into account that for ice going ships damages of shaft line is quite routine due to high asymmetric loading on propeller when one or two blades operates in ice and another two or three in water CRP shaft system will be very vulnerable and non-reliable in ice. CRP shaft + pod is less complicated and in principle may be used for ice class ships, but its efficiency gain is much two lower for slow speed H:\MEPC\72\MEPC 72-INF.12.docx

260 MEPC 71/INF.12 Annex 13, Page 12 Categorization / Type of (2) Free rotating vane wheel Short-description (Revised by CESA) The vane wheel (Grim wheel) is a freely rotating propeller, installed behind the main propeller. The vane wheel has a larger diameter than the main propeller. The part that is directly behind the main propeller is turned by the swirl from that propeller and acts like a turbine, driving the part of the vane wheel that is outside the diameter of the main propeller. This outer part acts as a propeller. Losses in the axially accelerated flow behind the propeller and part of the rotational energy is thus transformed into propulsive energy. s in power consumption are reported to be around 10%. It is claimed that an important benefit of the vane wheel is that the smaller main propeller that can then be installed results in a lighter and less costly propulsion unit. The long and slender vane wings may be damaged at sea, especially in a heavy seaway. It should be noted that, if there is space in the after body for a vane wheel, there will also be space for a main propeller of larger diameter, offering approximately the same improvement in power consumption as the ratio (%) (Japan) 0-10% of power consumption (Japan) 10 percent would be too large. Therefore, 10 percent should be considered as maximum improvement ratio. ratio when using with other energy saving (%) slightly reduced. Combination with category 2(2) is not applicable. (Denmark) Will not work together with other post-swirl devices like PBCF, Costa Bulb, Asymmetric rudder (Japan) Perspective of application for ice class ships. ships and in general so complicated and expensive power plant is not commercially attractive. This case till now was not under investigation. There are only sample of CRP application on ice class ships Steerprop CRP pods on icebreakers with drive of both propellers by mechanic gear. Its power is limited and they are installed on icebreakers, where propulsion in open water is not of first order. Non-acceptable for ice class ships. Grim hill have big strength problems even for open water case (long narrow blades operated in propeller slipstream). Block of ice after the propeller will increase the problems to irresolvable. H:\MEPC\72\MEPC 72-INF.12.docx

261 MEPC 71/INF.12 Annex 13, Page 13 Categorization / Type of Short-description combination of a small main propeller and a vane wheel. The vane wheel should be a suitable potential improvement for cargo ships. ratio (%) ratio when using with other energy saving (%) This technology collect the energy of rotational flow. This is why this could interfere with other energy saving devices which have same effect such as pre-swirl devices and post-swirl devices. To fit large diameter propeller could cause negative affect to the improvement of hull form. Perspective of application for ice class ships. (CESA) can be combined with most of the other without loss of H:\MEPC\72\MEPC 72-INF.12.docx

262 MEPC 71/INF.12 Annex 13, Page 14 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) improvement ratio, in particular categories 1(1)(2), 4, 5 and 6(1)(2)(3). The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. Perspective of application for ice class ships. (3) Ducted propeller The ducted propeller consists of a propeller mounted centrally in a ring foil. Compared to the conventional propeller of the same diameter and thrust, this arrangement allows a larger mass of water to be supplied to the propeller, improving the operating conditions around the propeller and the ideal efficiency. The duct generates additional thrust. The potential for reduced power consumption on relevant ships has been reported to be in the range 5-20%, with perhaps 10%being a good average value. The duct results in increased resistance, but at higher propeller loadings this is more than compensated for by the positive effects of the combination of propeller and duct. Below is the diagram (Japan) 0-10% of power consumption (The improvement ratio depends on the thrust loading condition.) (CESA) can be combined with most of the other without loss of improvement ratio, in particular Ducted propellers sometimes are used on the ships of light and medium ice classes (such as tugs) to improve thrust at high loading, but it is not energy saving device. For ice ships quite routine is blockage of duct by ice blocks and it requires reversal of propeller rotation (or H:\MEPC\72\MEPC 72-INF.12.docx

263 Propeller efficiency (-) MEPC 71/INF.12 Annex 13, Page 15 Categorization / Type of Short-description that illustrates how the propeller efficiency depends on the thrust loading coefficient CTh (defined in 1. (3)) Ideal efficiency Conv. propeller (Wageningen B-series) Ducted propeller Propeller efficiency ratio (%) (CESA) Only useful for ship types with high propeller loading. ratio when using with other energy saving (%) categories 1(1)(2), 4, 5 and 6(1)(2)(3). The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. Perspective of application for ice class ships. reversal CPP blade position) to wash out the ice from duct. For high ice classes propeller without duct is applied Propeller thrust loading coefficient Cth (-) The ideal efficiency is a purely theoretical efficiency assuming no losses. The curves are based on the previous diagram by Andersen and Breslin and show that only for ships with a CTh value higher than 2 ducted propellers can be recommended. It could also be mentioned that some propeller manufacturers have improved the efficiency of ducted propellers by further optimization of the duct, such that the propeller thrust at low speed (bollard pull condition) has been increased by 6 8 %.Ducted propellers are therefore suited for ships operating at high propeller loadings, such as tankers, bulk carriers, tugs and different offshore supply and service vessels. H:\MEPC\72\MEPC 72-INF.12.docx

264 MEPC 71/INF.12 Annex 13, Page 16 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. (4) Pre-swirl devices These are devices that aim to provide a favourable pre-rotation of the flow of water in front of the propeller. They include radial reaction fins in front of the propeller and an asymmetric stern. Considering radial reaction fins, a reduction in power consumption has been given as 3-8% from tests with models, while the result that has been reported from a full-scale test was 7-8%. For the asymmetric stern, improvements in power consumption of 1-9%have been reported in tests on models. The improve ratio differs according to the ship type or the propeller characteristics and so on. Radial reaction fins or an asymmetric stern should be applicable to all single-screw ships, and should work according to expectations in many cases. It should be noted, however, that in many cases the expected benefits have not been demonstrated in full-scale operation. Also, there are two examples in relation to Pre-swirl devices, namely, Guide vanes in front of the propeller and Wake-equalizing duct should be integrated in-use. Example1Guide vanes in front of the propeller The aim of guide vanes is to eliminate or reduce the cross-flow that is often observed in front of the propeller. These vanes are fitted in front of the propeller on both sides of the sternpost. The vanes straighten the flow in the boundary layer in front of the propeller, thereby improving its efficiency. Cross-flow appears mostly in ships with stern bulbs and full hull forms that operate at relatively low speed. The benefit is therefore largest for tankers and bulk carriers. The improvement decreases with decreasing fullness of the hull form. Example2Wake-equalizing duct (Japan) 1-4% improvement of the power consumption ( 4-6% improvement would be achievable in case post-swirl device in 2.(5)) is used together) (Japan) The improvement ratio is reduced because of interfere of devices. (CESA) Full scale tests on an optimised PSS indicate 7% improvement (PD) for a bulk carrier (ref. EU (Japan) This could interfere with CRP or other pre-swirl devices. The effect of this technology depends on stern hull form. This device collects stern viscous vortex, and this is why the effect is larger for fat ships than slender ships. (CESA) 8% improvement would be achievable in case post-swirl device in 2.(5)) is used together can be combined with Pre-swirl devices is not acceptable for ice class ships because even for light ice classes there is a danger to cut off pre-swirl devices by ice. The danger is higher because some portion of hull plating may be cut off together with devices. So, such devices are not applicable for safety reasons. H:\MEPC\72\MEPC 72-INF.12.docx

265 MEPC 71/INF.12 Annex 13, Page 17 Categorization / Type of (5) Post-swirl devices Short-description The wake-equalizing duct consists of one half-ring duct with foil-type sections attached on each side of the after body, in front of the propeller. The half-ring duct accelerates the flow into the propeller in the upper quadrant on each side and retards the flow in the lower quadrants. This results in a more homogeneous wake field in front of the propeller, while the average wake is almost unaltered. The improved power consumption that is obtained from well-designed wake-equalizing ducts results from several component savings: improved efficiency because of more axial flow and a more homogeneous wake field; reduced resistance because of reduced flow separation at the after body; lift on the ducts directed forward; orientation of duct axes so that the inflow to the propeller is given a small pre-rotation; and improved steering, due to straightened flow over the rudder and more lateral area aft. A number of devices belong to this category. Several of them involve modifications to the rudder. The most important among these devices may be additional thrusting fins at the rudder, rudder bulb systems with fins, fins on the propeller fairwater (boss cap fins) and an asymmetric rudder. For these devices, improvements in power consumption of 1-8% have been reported from tests on models. From full-scale measurements, an improvement of 8-9% has been measured for ratio (%) GRIP project, ww.sciencedire ct.com/science/ article/pii/s ) 1-4% improvement of the power consumption ( 4-6% improvement would be ratio when using with other energy saving (%) most of the other without loss of improvement ratio, in particular categories 1(1)(2), 4, 5 and 6(1)(2)(3). The improvement potential of combinations within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. (Japan) This could interfere with CRP or preswirl devices. The effect of this technology Perspective of application for ice class ships. No post swirl devices, which have a different kind of foils on the rudder, can be applied on ice class ships because foils will be cut off by ice. Integrated rudder scheme and asymmetric rudder also are not H:\MEPC\72\MEPC 72-INF.12.docx

266 MEPC 71/INF.12 Annex 13, Page 18 Categorization / Type of Short-description additional thrusting fins at the rudder, while 4% has been reported for boss cap fins. Post-swirl devices should be applicable to all new ships, but, as for pre-swirl devices, the benefits, in many cases, have been difficult to demonstrate in full-scale operations. Also, there is an example in relation to Post-swirl devices, namely, integrated propeller and rudder units should be integrated in-use. Example 1 Integrated propeller and rudder units As the name implies, the propeller and rudder are designed as an integrated unit, part of the design being a bulb behind the propeller that is fitted into the rudder. At least two patented designs exist. The effect of these units has been reasonably well documented in tests on models and in full-scale trials. An improvement of 5% in power consumption may be taken as typical. The units are applicable to general cargo vessels, RoPax vessels and container vessels operating at relatively high speed. ratio (%) achievable in case pre-swirl device in 2(4) is used together) ratio when using with other energy saving (%) depends on stern hull form. (CESA) 8% improvement of power consumption would be achievable in case pre-swirl device in 2(4) is used together (CESA) can be combined with most of the other without loss of improvement ratio, in particular categories 1(1)(2), 4, 5 and 6(1)(2)(3). The improvement potential of combinations Perspective of application for ice class ships. acceptable because rudder of ice class ships should operate in ice safely and effectively both in forward and astern mode. As exception one can consider boss cap fins with short strong fins on the propeller hub cap, but their efficiency gain is not higher than 2% (this device utilize the hub vortex energy loss only. In the case of podded propulsor when propeller operates in front of pod, application of such device is impossible. H:\MEPC\72\MEPC 72-INF.12.docx

267 MEPC 71/INF.12 Annex 13, Page 19 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) within category 2 and with 7(3) might be slightly reduced. Combination with category 2(2) is not applicable. Perspective of application for ice class ships. (6) Split stern (Twin skeg) Split stern has effect on improvement of propeller efficiency caused by mitigation of propeller loading, and is suitable to beamy ship. 0-6% (CESA) Potential in application highly depends on the application case. (CESA) can be combined with most of the other without loss of improvement ratio, in particular categories 1, 4, 5, 6 and 7. The improvement potential of combinations within category 2 and with 1(1), 7(1) might be As high and medium ice class ships should have reinforced bossings for shaft protection, kind of split stern in fact is conventional for ice class ships, but it is never considered as energy saving measure. H:\MEPC\72\MEPC 72-INF.12.docx

268 MEPC 71/INF.12 Annex 13, Page 20 Categorization / Type of Short-description 4 Engine energy recovery (1) for lowspeed engine producers of low-speed engines for many years. The first system offered Energy-recovery systems for ship engines have been available from was usually based on an additional exhaust turbine which was fed from the exhaust receiver by a fraction of the engine exhaust flow, in the range of 10%. The exhaust turbine could be connected to the engine's crankshaft or alternatively to an electric generator. Since fuel prices in the past have been too low to make these systems profitable, the number of installed systems is relatively small. For some years, more advanced systems have been developed and are today commercial, at least for low-speed engines. An example is B&W's system TES (thermo efficiency system), which combines a turbine in the exhaust gas with a steam cycle that is driven by exhaust heat and running a steam turbine. The two turbines are coupled to a generator for production of electrical power. The power can then be used to drive a shaft generator/motor to assist the main engine, or consumed elsewhere in the ship. The corresponding increase in engine power is estimated to being the range of 9 to 11%, which, in terms of shaft efficiency, increases to about 55% (from about 49.5%). The contributions from the two systems are respectively 5% and 6%, from the exhaust turbine and the steam turbine. The efficiency of the steam cycle is somewhat limited by the minimum recommended temperature of the exhaust stack, which must be above 180 C to control the formation of deposits and the corrosion by sulphur oxide that are related to the use of fuel oils. ratio (%) (Japan) Waste Heat Recovery System could recover 2-6% reduction of power consumption. (Denmark) 6-12% seems high 3-6% is more realistic (Japan) The effect of this device could be taken into account as reduction of PAE. Maximum effect of this would be about 5%. The effect of this technology would be smaller if ratio when using with other energy saving (%) slightly reduced. (CESA) can be combined with almost all of the other without loss of improvement ratio. Combination with category 4(2)(4) is not applicable. Perspective of application for ice class ships. The using of engine energy recovery systems is difficult and ineffective for ice class ships due to the following reasons: 1. During the operation of the ship in ice, there is a need for frequent reversals and continued operation of the main engines on fractional loads. As a result, steam generation in the waste heat recovery boilers is unstable, fires in the waste heat recovery boilers due to soot sediments, increased lowtemperature sulfur corrosion. Due to the high probability of accidents in the exhaust steam boilers on ice navigation vessels, there is currently a tendency of H:\MEPC\72\MEPC 72-INF.12.docx

269 MEPC 71/INF.12 Annex 13, Page 21 Categorization / Type of (2) for mediumspeed engines Short-description Steam cycles, as a means of energy recovery, have some properties that are quite challenging on board a ship. The relatively low temperature level makes systems relatively bulky. In particular, the condenser operates at the low steam density that exists at the actual condensation temperature. There are some interesting forthcoming developments that are expected to make a significant impact on the gain in engine efficiency. Organic Rankine Cycle systems have been designed and are already commercial. They show some favourable properties, in particular much smaller space requirements compared to a steam system. The working fluid is currently alkanes or refrigeration fluids. Due to the properties of the working fluids (fire hazard, ozone-depletion properties), highpressure CO2 is considered to be a more desirable working fluid. The medium-speed diesel engines have a lower fuel efficiency compared to low speed ones, usually in the range 42% to 44.5%. These ratio (%) temperature of exhaust gas is relatively low. 3.5% of shaft power 5% by turbo Generator; and 10% by turbo Generator + Power Turbine (Denmark) 3,5 % seems high (Japan) The effect of this device ratio when using with other energy saving (%) (CESA) can be combined with almost all of the other without loss of improvement ratio. Combination with category 4(1) is not applicable. Perspective of application for ice class ships. using thermal oil instead of steam. 2. For ice navigation uses a lot of heat for heating fuel and ballast tanks, as well as other support needs. This circumstance substantially reduces the possibility of using steam in a turbo generator. Same as for low-speed engine H:\MEPC\72\MEPC 72-INF.12.docx

270 MEPC 71/INF.12 Annex 13, Page 22 Categorization / Type of Short-description engines normally have an exhaust gas temperature in the range of 300 C to 360 C. While the minimum temperature of the exhaust stack is 180 C or above, which poses a limitation to heat recovery, the energy utilization is calculated to be in the range of about 3.5% of the shaft power. ratio (%) could be taken into account as reduction of PAE. Maximum effect of this would be about 5%. The effect of this would be smaller if temperature of exhaust gas is relatively low. ratio when using with other energy saving (%) Perspective of application for ice class ships. H:\MEPC\72\MEPC 72-INF.12.docx

271 MEPC 71/INF.12 Annex 13, Page 23 Categorization / Type of (3) for gasfuelled engines Short-description Due to increasing prices of fuel oil and regulations to control exhaust emissions (restricting operations or taxation), there is an increasing interest in using gas engines, burning natural gas, in marine applications. The currently available gas-fuelled engines for ship propulsion, with piston bores in the range of 25 cm to 50 cm, have slightly higher shaft efficiency(at MCR) compared to their diesel counterparts, normally in the range of 44.5% to 47%depending on engine size and engine concept (spark-ignited and dual-fuel). The gas engines offer a higher potential for energy recovery. This comes from the higher exhaust temperatures (normally in the range of 400 C to 430 C) and a possibility to run with an exhaust stack temperature below 100 C. This can be done, while the fuel does not contain any sulphur and the combustion produces very small quantities of particles. Thereby, the exhaust system is likely to be little affected by the low temperature of the exhaust. ratio (%) Reduction of CO2 22% (including 13% energy shaft power recovery) 5% by turbo Generator; and 10% by turbo Generator + Power Turbine ratio when using with other energy saving (%) (CESA) can be combined with almost all of the other without loss of improvement ratio. Combination with category 4(1)(2) is not applicable.. Perspective of application for ice class ships. Same as for low-speed engine except problems with lowtemperature sulphur corrosion. A simplified calculation, based on an exhaust gas temperature of 430 C and an exhaust stack temperature of 50 C, gives a theoretical (Carnot) efficiency of about 32.5%. Considering a turbine efficiency of 80% and some additional losses related to pumping and heat exchangers, the total recovery efficiency could be in the level of 22%. Using actual figures for exhaust heat flow and engine shaft power, the energy-recovery figure is 13% of the shaft power. That means that the actual shaft efficiency increases from 45% to 50.9%. (Japan) <Comment: Clarification is necessary what kind of gases are taken into account in this explanation.> H:\MEPC\72\MEPC 72-INF.12.docx

272 MEPC 71/INF.12 Annex 13, Page 24 Categorization / Type of (4) Machinery arrangement and hybrid propulsion Short-description Currently, tankers, bulk carriers, containerships and general cargo ships have one large low-speed propulsion engine directly connected to the propeller. This arrangement has proven to be very efficient and, since the ships operate mainly at high engine load, there is little to gain by complex multi-engine machinery arrangements or by using hybrid propulsion systems. For the RoPax/cruise segment, it is currently common to use multiple engines and two or more propellers. A primary reason for this is restrictions on draught and high power demand. An additional reason is the space restrictions and the use of medium-speed engines; hence there is already a need (and transmission loss) for a gearbox. This multi-engine situation opens up some possibilities for designing systems that can handle variable loads. There are some applied for certain types of ships. Today most super yachts have geared diesel driven twin screw propulsion with open shaft lines and twin spade rudders. However a trend towards hybrid propulsion has emerged. The main reason for this is the varied operating profile of super yachts and the consequences thereof for propulsion efficiency. A recent survey of a sample of 27 yachts has shown that over 50% of the operating time was at less than 20% of engine load. Less than 5% of the operating time was over 80% engine load. Annual average operating hours were as low as 277 per yacht (MEPC 65-INF.15). Such operating profiles differ considerably from those of most merchant vessels. Hybrid propulsion systems enable higher propulsion efficiencies at all engine loads. (CESA) ratio (%) CRP-Pod vessel could improve the propulsion performance by 13% compared with the conventional twin shafts and propellers system. (Japan) The effect of this technology would be lower. ratio when using with other energy saving (%) (Japan) The effect of this would interfere with pre-swirl and post-swirl device (CESA) can be combined with almost all of the other without loss of improvement ratio. Combination with category 4(1) is not applicable. Perspective of application for ice class ships. 1. In principle, a hybrid propulsion systems can be used for ice class ships, especially in the version of the afterburner to increase the power when the vessel is operating in ice. But this option does not lead to a decrease in EEDI, since in the EEDI calculations the total power is used. 2. CRP-Pod technology cannot be used for ice class ships, since there is a high probability of the propellers damage in ice (see the explanations above - (4) and (5)). H:\MEPC\72\MEPC 72-INF.12.docx

273 MEPC 71/INF.12 Annex 13, Page 25 Categorization / Type of Short-description <Comment: this goes beyond the short description of technology; contains information that could be used when considering the potential of combined > ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. Information for the Hybrid propulsion (diesel-electric propulsion combined with use of batteries for energy storage): Most diesel engines have their optimum performance (lowest specific fuel oil consumption, SFOC) at % MCR (if not specially tuned for low load operation where the lowest SFOC occurs at a lower MCR percentage). Some diesel-electric propulsion systems operates at a variable load situation which means that multiple gen-sets occasionally operates at very low load, i.e. far away from the optimum point resulting in a high SFOC and sometimes at a very high load close to 100 % MCR. For such systems it is beneficial to change the operation profile such that all gen-sets are operated at the optimum point, such that the surplus of electric power is accumulated in batteries for use in the periods where there is a need for high propulsion power. With such a so-called hybrid system often one or two gen-sets can be omitted, as the extra electrical is taken directly from the battery package. Hybrid systems are comfortable on diesel-electric ferries on short routes where there is a very high power demand during the acceleration phase, while the power demand is limited when the service speed has been achieved. (5) Fuel-cell propulsion (Revised by CESA) Fuel cells have high potential thermal efficiency and low emissions. For this reason, fuel-cell technology is, in principle, also an interesting alternative to the use of traditional combustion engines for merchant shipping. Fuel cells can be used either as standalone or in a combined cycle, where exhaust heat is recovered for additional generation of (CESA) (5-10%)* 25-30% (100%) (CESA) can be combined with all other without loss of In principle, fuel-cell propulsion systems can be used for ice class ships. But there is problem of bunkering if for fuel-cells are used non- H:\MEPC\72\MEPC 72-INF.12.docx

274 MEPC 71/INF.12 Annex 13, Page 26 Categorization / Type of Short-description power. Fuel-cell systems have been identified as particularly promising power generators for both ship hotel power and also for hybrid propulsion systems, where they work in combination with a diesel engine. There are many issues relating to the use of fuel cells on board ships. Fuel cells use non-conventional fuels, such as hydrogen, methanol, lowflashpoint diesel, some requiring fuel preparation (reforming). With decreasing price and increasing reliability, fuel cells presently become more interesting than before. Main technological obstacles to operating fuel cells on board medium to large ships have been removed and international safety standards are in process of finalization. The related R & D projects have reached the implementation stage with full scale on-board trials, including: development of fuel processing systems for fuel-cell units capable of running on liquid fuels; energy-recovery systems (e.g., boilers, turbines) for use in conjunction/integration with high-temperature fuel-cell systems (MCFC and SOFC); standardization of fuel-cell systems (including auxiliary systems) into modules of 0.5 MW to 1.0 MW size; intrinsically safe systems for on board storage of fuel and fuel handling; and development and full-scale validation of systems with respect to their use in the marine environment: reliability, availability, vibration, accelerations, salinity, humidity, and ability to respond to transient power demands. ratio (%) * aux. engines only in short sea shipping with hydrogen produced with renewable energy ratio when using with other energy saving (%) improvement ratio.. Perspective of application for ice class ships. conventional fuels (such as hydrogen or methanol). If for fuel-cells are used conventional fuels with reforming the improvement ratio (%) in comparison with the diesel engine raises doubts. (Denmark, CESA) In cogeneration (Combined Heat and Power, CHP) applications, a conventional power plant producing electricity is enhanced with a heat 30% (VFD) more efficient WHR plus VFD = 5% increase The using of engine energy recovery systems is difficult H:\MEPC\72\MEPC 72-INF.12.docx

275 MEPC 71/INF.12 Annex 13, Page 27 Categorization / Type of (6) Boiler efficiency with Waste Heat Recovery (Economizer) Boiler configuratio n Desig n with burne r havin g VFD plus damp er Adding a pre- heater (waster heat recovery in conjunction with a boiler (Denmark) (7) Optimization of Exhaust gas via Waste Heat Recovery ( Main & Aux engine) Short-description recovery system to utilize the energy otherwise wasted in the exhaust gas. Since the exhaust gas waste heat is a free source of energy; the more you recover - the more you save. Replacement of the damper air control with a Variable Frequency Drive can improve the fuel usage for the ship. VFD provides electrical savings that can be due to oversize fans, oversize motors, and minimizing damper horse power due to lower damper pressure drop. Energy saving could vary between 5 to 30% depending on boiler firing rate. A Waste Heat Recovery Unit can be added to the boiler exhaust so that latent heat exhaust can be used to preheat the feed water. Increasing heating efficiency by reducing the amount of energy needed for heating the water to create steam. Additionally, when the WHR is used together with a condenser the flue gases are cooled down to condense water vapour from burning of hydrogen, allowing extra energy recovery. The WHR coupled with boiler will increase boiler efficiency by a further 10% to 15%. Waste heat recovery systems recover the thermal energy from the exhaust gas and convert it into electrical energy, while the residual heat can further be used for ship services (such as hot water and steam). The system can consist of an exhaust gas boiler (or combined with oil fired boiler), a power turbine and/or a steam turbine with alternator. Redesigning the ship layout can efficiently accommodate the boilers on the ship to better fit these systems. ratio (%) at a firing rate of 80% ratio when using with other energy saving (%) in efficiency (30% +5%) 3.9 to 5.6% Up to 15% when used in conjunction with a condenser Perspective of application for ice class ships. and ineffective for ice class ships due to the reasons same as for low-speed engine (1). The using of engine energy recovery systems is difficult and ineffective for ice class ships due to the reasons same as for low-speed engine (1). H:\MEPC\72\MEPC 72-INF.12.docx

276 MEPC 71/INF.12 Annex 13, Page 28 Categorization / Type of Short-description In engine systems much of the energy produced by the engines goes to heat rather than power - 25% of an engine s fuel energy goes into the exhaust flow as waste heat. By recovering this energy with a boiler or waste heat recovery system (WHR) it is possible to reduce both fuel consumption and emissions. The converted energy can be used for production of steam and hot water, production of extra electricity, heating of fuel oil and/or supplying district heating and cooling. The waste heat from the auxiliary engines has not been considered in the past, but it contains a large amount of energy which can be utilized to supplement steam requirements during port stays and - for some vessels - also during voyage. Unlike the continual operation of the main engine during oceangoing voyages, the operation of the auxiliary engines varies. The WHR, therefore, has been developed as a customized solution focused on generating energy under varying load conditions. To ensure the most advantageous design, the Aalborg WHR will be specially tailored to the individual ship and engine design with due consideration to the existing uptake back pressure and other critical factors The WHR concept has been developed as a customized solution with special focus on energy generation compared to return on investment and has a very short payback time of down to 4-6 months in optimum cases, but normal payback time will be approximately 1 to 1½ year depending on the number of days the produced steam can be utilized (offset against the steam requirement from the oil-fired boiler) and the redundancy requirements. ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. When the WHR is used together with a condenser the latent heat of the water vapor is recovered. The lower the water temperature, the greater the amount of energy recovered. H:\MEPC\72\MEPC 72-INF.12.docx

277 MEPC 71/INF.12 Annex 13, Page 29 Categorization / Type of Short-description Overall, Lowering the fuel oil consumption on the main engine, through increased efficiency therefore results in higher fuel oil consumption of the oil-fired boiler if the entire steam boiler plant remains unchanged ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. (CESA) (7) Exhaust after treatment solutions (SCR) 5 Hull coatings (1)Selection of coatings SCR-Technology can be used for Tier II and for Tier III NOx applications. Especially when operated in conjunction with a Tier II requirement, the NOx reducing performance of a SCR-system can be utilized in a way that the engine is to be optimized to the best possible specific fuel consumption (SFC), accepting elevated engine NOx out emissions, which are reduced to the required Tier level by use of a SCR-system. Optimized SFC results in reduced CO2-emissions. In a research project managed by MARINTEK, a number of coatings, of both anti-fouling and foul-release type, from five major manufactures have been tested on 16 Norwegian ships over a period of seven years. The results show that these new coatings are equally as effective astbt-based systems. (TBT-containing paints are banned as a result of the International Convention on the Control of Harmful Anti-fouling Substances on Ships, 2001 established by IMO.) 3-5% Can be combined with other 1-2% reductions of frictional resistance by Low friction coating (Japan) The effect of this could be taken into account as the difference of Vref at sea trial. This is why the effect to attained EEDI is only 1-2 %. (Japan) The effect would be reduced when using other friction reducing devices (CESA) can be combined with all other without loss of improvement ratio. The improvement May be used for ice class ships. Acceptable for ice class ships the coating, which can resist to interaction with ice. H:\MEPC\72\MEPC 72-INF.12.docx

278 MEPC 71/INF.12 Annex 13, Page 30 Categorization / Type of Short-description ratio (%) ratio when using with other energy saving (%) potential of combination with 5(2) might be slightly reduced. Perspective of application for ice class ships. (2)Polymers and air lubrication Frictional resistance can be reduced by modifying the wetted surface of the hull, such as by introducing riblets that mimic shark scales or by applying an artificial enhancement (such as the use of air bubbles and/or air cavities and polymers). Research is still going on concerning air lubrication on hull forms for conventional ships, but it has so far not provided significant improvements. Air-lubrication technology is claimed to provide reductions in resistance that are in excess of 5%, which is significant in this context. Adding a small amount of polymer to a turbulent Newtonian fluid flow can result in a reduction of the viscous frictional resistance. During the past three decades, numerous research activities were dedicated to the reduction of frictional resistance by applying polymers. As a result, roughly three main methods of friction reduction by polymers have been developed. The first method is based on a molecular scale, due to the fact that the behaviour of polymer molecules in various model flows has been studied. The second type of method relies on investigation of the effects of polymers on the time-averaged turbulence statistics, while the third type of method examines changes in the coherent turbulent structure due to the presence of polymers. As in the previous case of air-lubrication technology, the three methods of using polymers to reduce frictional resistance are not yet mature, i.e. research in that direction is still going on. Additionally, the concept of continuously injecting polymers into the water may not be suitable for sustainable (Japan) 1-3% (The improvement ratio is different according to the ship type; higher improvement ratio could be expected for a blunt ship such as a tanker and a bulk carrier, than a slender ship such as a container carrier.) (Japan) The effect of air lubrication would be reduced if flat part of bottom (Japan) The effect would be reduced when using other friction reducing devices (CESA) can be combined with most of the other without loss of improvement ratio. The improvement potential of combination with 5(1) and 7(1) might be slightly reduced. Air lubrication is patented and tested on some of ice class ships for reduction of ice friction against the hull in ice operation. In principal air lubrication may be considered for ice class ships also as energy saving device. Hull coating is acceptable if coating is able to resist to interaction with ice. Polymer supply to the flow around the hull is very specific problem which can be considered at the moment as scientific exercise. Small effect of friction reduction is more or less approved. But it might be anticipated a lot of problems with storage of polymer on the ship, special system for preparation of polymeric solution to supply it through the pipe to water. Experience on polymer supply in ice is H:\MEPC\72\MEPC 72-INF.12.docx

279 MEPC 71/INF.12 Annex 13, Page 31 Categorization / Type of Short-description operation. Therefore, the concept of polymer injection is not considered to be very important for reduction of ship resistance. However, it should be noted that any improvements to the wetted surfaces of the hull that are achieved by these means may also inhibit organic growth. None of the mentioned are proven in service. Additionally, an air-bubble system would require energy to produce the bubbles. Hull coatings based on nanotechnology have been advertised by different companies for some time now, and have also been mentioned in the media recently. It is claimed that these coatings have the potential of reducing the basic viscous frictional resistance of the underwater hull to a considerable extent and to delay the onset of marine growth for an extended period. The claims are largely unsubstantiated at present, but, if they can be even partly realized in the future, power reductions of perhaps 15% may be expected. Thus, this type of coating of the underwater hull will be one of the most important contributions toward reducing fuel consumption and CO2 emissions for well-designed conventional ships. It will be particularly favourable that such coatings probably can be applied both to new ships and to existing ships. ratio (%) is narrow or depth is large. Deeper depth requires more energy for providing air to the bottom. ratio when using with other energy saving (%) Perspective of application for ice class ships. completely absent. But according to experience to be active each hole in the hull of the ship operated in ice should have warming system. Polymer supply to water from a lot of ship inevitable will have effect on the mammal, which filtered water (jelly, plankton, etc.) (1)Liquefied natural gas (Revised by CESA) Gas that is stored in the liquid state, as liquefied natural gas (LNG), is predicted by many as a forthcoming fuel for ships. Key drivers for this expected development are the low emissions of nitrogen oxides (NOx), SOx and particulate matter (PM) from LNG-fuelled ships. Also, LNG contains more hydrogen and less carbon than diesel fuels; hence emissions of CO2 are reduced. Unfortunately, increased emissions of methane (CH4) reduce the net effect. The price of LNG is significantly (Japan) 20% (CESA) 20%-25% (CESA) can be combined with all other without loss of improvement. May be used for ice class ships but there is problem of LNG bunkering as a result of the lack of infrastructure in the northern regions. H:\MEPC\72\MEPC 72-INF.12.docx

280 MEPC 71/INF.12 Annex 13, Page 32 Categorization / Type of Short-description less, compared to distillate fuels; therefore there is a considerable economic incentive for a move towards using LNG. The most important technical challenge is finding the necessary space for storage of the fuel on board the ship and the availability of LNG in the bunkering ports. Therefore, LNG is primarily interesting in a regional shipping context, where the ship's range is less of an issue and the next port of bunkering is more predictable. LNG could also become an interesting fuel for tankers, since there is considerable space available for the LNG fuel tanks on deck. LNG-fuelled ships would be particularly attractive in NOx emission control areas, since they can meet Tier III emission levels without after treatment of the exhaust gases. ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. LNG-fuelled ships can use either pure gas-fuelled engines or dual-fuel engines that are capable of burning gas, diesel or combinations of these. LNG is a proven technical solution, with10 ships already in operation and 19 ships on order. Currently, the cost of bulk LNG is about the same as that of residual (heavy) fuel oil, and significantly cheaper than distillate (fossil) fuels. Natural gas can also be processed to produce Fischer Tropsch diesel, for use in diesel engines; however, in this case, the NOx benefit that is associated with LNG operation would be lost. Also, natural gas can be reformed on site and used as fuel for fuel cells; however, this is currently not considered to be an interesting option due to the principal fuel-cell challenges (including cost, durability and power density). Presently, only four-stroke medium-speed engines for direct-drive LNG propulsion are already on the market. (2)Biofuels These fuels include current, "first-generation" biofuels made from sugar, starch, vegetable oil or animal fats, using conventional technology. Among these, biodiesel (i.e. Fatty Acid Methyl Esters, FAME) and (CESA) can be combined with May be used for ice class ships. H:\MEPC\72\MEPC 72-INF.12.docx

281 MEPC 71/INF.12 Annex 13, Page 33 Categorization / Type of Short-description vegetable oils can readily be used for ship diesel engines. In rough terms, biodiesel could substitute distillate fuels and vegetable oils could substitute residual fuels. With some biofuels, there may be certain issues such as stability during storage, acidity, lack of water-shedding, plugging of fuel filters, wax formation and more which suggest that care must be exercised in selecting the fuel and adapting the engine. Blending bio-derived fuel fractions into diesel or heavy fuel oil is also feasible, from the technical perspective; however, compatibility must be checked, as is also the case with bunker fuels. Future processes to convert biomass into liquid fuels can be designed to synthesize various fuels that are suitable for use on board ships. Currently, biofuels are significantly more expensive than oil-derived fuels. This would have to change if there is to be an incentive to use these fuels on board ships. Moreover, as discussed in the future scenarios, as long as there is a demand, driven by legislation, for biofuels to be used and for carbon reductions on shore, it will be natural to preferentially use biofuels on land, where this is credited, rather than on ships. ratio (%) ratio when using with other energy saving (%) all other without loss of improvement. Perspective of application for ice class ships. (CESA) (2bis) LPG (Japan, CESA) (2ter) Alcohol Liquefied Petrol Gas (ref. (Methanol) Same as liquefied natural gas (LNG), methanol is considered as one of alternative fuels which could reduce environmental pollution to the % can be combined with most other without loss of improvement ratio (see comments for LNG) Methanol 7-10% can be combined with most other May be used for ice class ships but there is problem of LNG bunkering as a result of the lack of infrastructure in the northern regions. May be used for ice class ships but there is problem of LNG bunkering as a result of H:\MEPC\72\MEPC 72-INF.12.docx

282 MEPC 71/INF.12 Annex 13, Page 34 Categorization / Type of Short-description atmosphere and the ocean. Like LNG, it does not contain sulphur, and this could contribute reducing particulate matter (PM) and NOx emissions. Regarding carbon dioxide (CO 2), methane (CH 3 OH) contains much hydrogen, like methane (CH 4) which is the main component of LNG, and this can reduce 10% emissions compared with heavy fuel oil. Methanol has good flammability, but it is difficult to ignite. For dual fuel low-speed diesel engines, the pilot fuel is injected to ignite and explode methanol fuel. Since a small amount of pilot fuel oil is used, 99% of PM is reduced, and up to 30% of NOx is reduced compared with heavy fuel oil. In comparison with LNG, the LNG fuel system requires ultra-low temperature technology, whereas methanol is liquid at normal temperature, easy to handle such as transportation and storage. Initial cost for methanol is about 1/3 of the cost for LNG. ratio (%) Ethanol 3-5% ratio when using with other energy saving (%) without loss of improvement ratio (see comments for LNG) Perspective of application for ice class ships. the lack of infrastructure in the northern regions. (Methanol) (ref. Effect to CO2- Emissions, Reference to DMX 1 Diesel /Gas Oil (DMX) Light Fuel (LFO) + 1 to + 3 % 3 Heavy Fuel Oil (HFO) + 2 to + 4 % 4.1 Liquefied Petroleum Gas (LPG) Propane - 10 to - 15 % 4.2 Liquefied Petroleum Gas (LPG) Butane - 10 to - 15 % 5 Liquefied Natural Gas (LNG) - 20 to - 25 % H:\MEPC\72\MEPC 72-INF.12.docx

283 MEPC 71/INF.12 Annex 13, Page 35 Categorization / Type of Short-description 6 Methanol - 7 to - 10 % 7 Ethanol - 3 to - 5 % ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. The effect of different fuel types to the calculation of the attained EEDI can be ascertained in the following way (ref. MEPC.281(70): The energy content (lower calorific value, LCV) of fuels, as proxy for the specific fuel consumption (SFC) of engines, is correlated to a reference fuel such as DMX (ISO 8217) and multiplied with the CF-value which stands for the carbon content on a mass-basis of the fuel concerned. The results represent the effect {%} of the different fuel types to CO2-emissions in correlation to the reference fuel DMX. (+) elevated CO2 emissions, (-) reduced CO2 emissions. The mechanical efficiency is assumed to be comparable for engine. (3)Wind power Wind power can be utilized in various ways on ships. These include:.1 Traditional sails;.2 Solid-wing sails;.3 Kites; and.4 Flettner-type rotors. Although sails were once the only source of propulsion, sails are currently considered to be interesting for providing additional supplementary power, as is suggested by recent studies, for instance. The use of traditional sails will impose bending moments to the hull, causing ships to list. Strength issues could result in a need for masts to run down to the keel, and the presence of the mast and rigging could have significant impacts on cargo handling. Kites differ from other concepts of wind power by having a small footprint during installation and hence being quite feasible to retrofit. Drawbacks with the kite systems include the complex launch, recovery and control systems that (CESA) 5%-15% [100%] for ship types with available deck space (CESA) can be combined with some of the other without loss of improvement ratio, in particular categories 5, 6, and 7. The improvements potential of of It is difficult to use wind power on ice-class vessels, because as a result of the ice generation equipment for the using of wind becomes inoperable and there is a high risk of stability loss. H:\MEPC\72\MEPC 72-INF.12.docx

284 MEPC 71/INF.12 Annex 13, Page 36 Categorization / Type of Short-description are needed. Also, the durability of the lightweight materials that are needed for kite sails is a challenge. Wing sails are solid structures resembling aircraft wings, which provide more thrust with less drag than conventional sails. Flettner-type rotors generate thrust from a rotating object in wind, taking advantage of the so-called Magnus effect. These systems have different characteristics with regards to how the thrust that is generated relates to other parameters, such as wind angle, wind strength, wind stability and ship speed. The energy of the wind varies by region and by area. In a study that was carried out at the Technical University of Berlin [18], three different types of sail were modelled onto two types of ships on three different routes. The objective of that study was to assess the savings of energy and of fuel that might be obtainable over a period of five years, using actual weather data. This study indicated that the potential for sail energy was better in the North Atlantic and North Pacific as compared to the South Pacific. Fuel savings were slightly larger at higher speeds; however, in terms of percentages; the fuel savings were greater at low speed due to the low total demand for propulsion power. In percentage terms, savings were typically about 5% at 15 knots, rising to about 20% at 10 knots. With optimal weather routeing, these figures improved. The best ship with the best sail type, with optimal weather routeing, operating in the most favourable five-year average weather (North Atlantic), was shown to save 15% at 15 knots and 44% at 10 knots. Presently, full-scale trials are being undertaken using kites. This technology is also discussed, in the context of marginal cost and abatement potential, in appendix 4. Naturally, it is difficult to simulate such complex systems, and currently there is limited full-scale experience with modern commercial sail ships against which such a model can be validated. Also, without such experience, it is also difficult to assess the practical feasibility of the size and number of sails modelled. The above figures should thus be ratio (%) ratio when using with other energy saving (%) category 1 and 7(1) might be slightly reduced. The engine / fuel related improvement potential remains, but applies to reduced installed power. Perspective of application for ice class ships. H:\MEPC\72\MEPC 72-INF.12.docx

285 MEPC 71/INF.12 Annex 13, Page 37 Categorization / Type of Short-description considered indicative. Nevertheless, sail-assisted power does seem to be an interesting opportunity for saving fuel in the medium- and longterm picture. ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. (4)Solar power A2.87 When assessing the potential of solar power for application on ships, it is interesting to consider the potential available energy. Earth's average solar irradiance on the surface is approximately 342 W/m 2. On average, 30% of this radiation will be reflected back to space. Clouds are the main contributor to the reflection. The solar irradiance will vary with latitude, season, weather conditions and time of day. How much of this energy a photovoltaic cell will be able to capture depends on the efficiency of the cell and the positioning of the cell relative to the solar beam. Current solar cells have an efficiency of about 13%. Today, the best technology, which is used in laboratories and on spacecraft, has an efficiency of approximately 30%. Efficiencies are predicted to reach 45 60% when third-generation photovoltaic cells are developed and matured. The specific power of solar cells is given in table. Table Indicative specific power of solar cells Current Current Future best Approximate energy conversion efficiency (%) Nominal power (W/m 2 ) Power adjusted for reflection(w/m 2 ) (CESA) 1% reduction of power consumption for ship types with sufficient deck space available (CESA) Could be combined with all other without reduction of the improvement ration; potential will increase in combination with (integration into) sails due to the increase of area available for panels The using of these is ineffective for ships of ice classes, since in the northern regions the solar activity is much lower than the average for the planet and long periods of the polar night in the northern regions. In addition, as a result of the generation of ice on solar batteries, they need heating to remove ice. Otherwise, there is a high risk of loss of stability. To get an idea of how much power it is possible to get from photovoltaic cells on a ship, the following example calculation has been made for a tanker with a length of 270 m and a breadth of 50 m (see table A2-4). A H:\MEPC\72\MEPC 72-INF.12.docx

286 MEPC 71/INF.12 Annex 13, Page 38 Categorization / Type of Short-description tanker of this size is equipped with an engine that is rated to approximately 18,000 kw, and the auxiliary power would be around 1,000 kw. ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. Table Power production by photovoltaic cells, assuming that the tanker's deck area is completely covered by solar cells Current Current Future best Approximate energy conversion efficiency (%) Nominal power (kw) 609 1,406 2,811 Power adjusted for reflection (kw) ,968 Current solar-cell technology would thus, on average, only be sufficient to cover a fraction of the auxiliary power even if the complete deck area was covered by photovoltaic cells. Therefore, it can be concluded that, due to the limited capacity of solar cells in respect to surface area that they cover, they do not yet appear to be a very efficient source of energy supply. Furthermore, at certain times and in certain areas, solar radiation will be above average and the auxiliary power demand could be met. Also, by using highly efficient (presumably expensive) spacecraft-type solar cells, current power demand could, on average, be met. Also, since solar power is not always available (e.g., at night) backup power would be needed; even if the power is available, on average, at day time, this would not help reduce the demand for auxiliary power at night unless there is an energy storage system available on board. H:\MEPC\72\MEPC 72-INF.12.docx

287 MEPC 71/INF.12 Annex 13, Page 39 Categorization / Type of Short-description Solar energy can also be used for heating purposes, e.g. of water while the ship is in port (Excess heat is normally available on board ships at sea.) (CESA) ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. 7 Other (1)Optimization Changing restriction of port, canal, etc. may bring the improvement of of principle propulsion performance by changing principle dimensions. e.g. Panama dimensions canal expansion. It depends on development of infrastructure. (CESA) can be combined with some of the other without loss of improvement ratio, in particular categories 5, 6, and 7. The improvement potential of of category 1 and 7(1) might be slightly reduced. The engine / fuel related improvement potential For ice class ships changing design restriction is very problematic because specific area of exploitation and strong influence of ice class rules. H:\MEPC\72\MEPC 72-INF.12.docx

288 MEPC 71/INF.12 Annex 13, Page 40 Categorization / Type of (2)Reduction of light weight (4)Hull Vane Short-description Increasing deadweight by reducing light weight, in keeping same displacement. A Hull Vane has been applied on a number of superyachts. The Hull Vane is a fuel saving device in the form of a fixed foil, located below the stern of a ship. The Hull Vane influences the stern wave pattern and creates hydrodynamic lift, which is partially oriented forward. This results in a reduction in of the ship's resistance. Refer to K. Uithof et al, "An Update on the Development of the Hull Vane R ", HYPER 2014 conference. Hull Vane has also been applied for supply vessels. Application for many other merchant vessels has proven to be effective on the basis of extensive CFD analyses. ratio (%) (Japan) 0-3% (Japan) The effect of H- CSR should also be considered. The effect of this technology depends on ship type and ship size. The improvement ratio depends on the ship types and the speed. It is reported that on merchant ships, the potential resistance reductions between 5% and 10% are common. ratio when using with other energy saving (%) remains, but applies to reduced installed power. (CESA) can be combined with all without loss of improvement ratio. (CESA) can be combined with some of the other without loss of improvement ratio, e.g. categories 1, 2, 4, 5 and6. Slightly reduced potentials can Perspective of application for ice class ships. All foils fixed on the hull will be cut of by ice, so this option is unacceptable for ice class ships. H:\MEPC\72\MEPC 72-INF.12.docx

289 MEPC 71/INF.12 Annex 13, Page 41 Categorization / Type of (5)Other energy saving applications on superyachts for Superyachts Short-description 10) Energy saving lighting such as LEDs. 11) Utilizing engine waste heat where possible for items like DPF catalyzers. 12) Engine selections and guidance to operators on speed ranges to minimize BSFC. These are verified on sea trials by running measured distance runs while recording actual fuel consumed. 13) Selection and design of underwater equipment to minimize drag, including stabilization equipment, cathodic protection devices, propeller strut shapes and alignment and propeller designs. 14) Tank testing and CFD work to maximize vessel efficiencies 15) Fitting of appendages such as bulbous bows for fuel economy 16) Extensive weight estimating and weight reduction strategies to increase vessel efficiencies (including application of aluminum and composite structural materials and lightweight interior materials). 17) Optimization of power management strategies to ensure that generators are producing power at maximum efficiencies. 18) Installation of Diesel Electric equipment to match operational profiles. ratio (%) ratio when using with other energy saving (%) be anticipated in combination with category 2. Perspective of application for ice class ships. Yachts designed for ice class should meet all the rules for ice class ships, so all notes presented above are actual for this point. (CESA) <Comments: Some of the above are also applicable to other ship types, e.g. 1) for all passenger ships and can be significant; H:\MEPC\72\MEPC 72-INF.12.docx

290 MEPC 71/INF.12 Annex 13, Page 42 Categorization / Type of Short-description Some of the above are already covered above and are applicable for all ship types, e.g. 4), 5), 6), 7), 8) and 9)> ratio (%) ratio when using with other energy saving (%) Perspective of application for ice class ships. *** H:\MEPC\72\MEPC 72-INF.12.docx

291 Annex 14, page 1 Effect of Rotor Sails on vessels EEDI coefficient Illustration of a Rotor Sail installation is shown only as an example, not related to cases presented H:\MEPC\72\MEPC 72-INF.12-annex 14 Confidential 08/01/2018 1

292 Effect of Rotor Sails on vessels EEDI coefficient Effect of Rotor Sails (Flettner rotor) on the EEDI number is presented Two example cases Deltamarin DWT bulker (B.Delta 210) Deltamarin DWT tanker (T.Delta 115) Effect on EEDI number is calculated according to MEPC.1/Circ.815 MEPC 72/INF.12 Annex 14, page 2 Perfomance values presented are annual averages based on average wind conditions (page 3 for further details) Thrust forces produced by Rotor Sails are calculated based on land tests and the verified performance results from Bore s M/V Estraden H:\MEPC\72\MEPC 72-INF.12-annex 14 Confidential 08/01/2018 2

293 Simulation parameters MEPC 72/INF.12 Annex 14, page 3 B.Delta 210 Total propulsion efficiency p=0.76 Effective thrust height from sea level 38 m MCR kw Speed at 75% MCR V=14.75 kn T.Delta 115 Total propulsion efficiency p=0.76 Effective thrust height from sea level 34 m MCR kw Speed at 75% MCR V=14.77 kn Power(main propulsion equivalent) = Thrust(RotorSail) x Service speed Total propulsion efficiency H:\MEPC\72\MEPC 72-INF.12-annex 14 Confidential 08/01/2018 3

294 Wind data used in EEDI simulation MEPC 72/INF.12 Annex 14, page 4 The wind data used in the simulation is based on IMO s MEPC 62/INF.34 (Marine Environmental Protection Committee) report Annex Global Wind Specification along the Main Global Shipping Routes to be applied in the EEDI calculation of wind propulsion systems. This wind specification is used for consideration of auxiliary wind propulsion systems in the EEDI (Energy Efficiency Design Index) and provides a good estimate for various routes. H:\MEPC\72\MEPC 72-INF.12-annex 14 Confidential 08/01/2018 4