Study: Draft Final Report

Transcription

1 AEBS Insert nd the LDWS title of your Exemptions presenttion Fesibility here Study: Drft Finl Report Presented by Nme Here MVWG Job Title Meeting, - Dte Brussels, 5 th July 2010

2 Contents Introduction Identifiction of Vehicle Types & Trget Popultions Vehicle stock, new registrtion, system cost nd effectiveness estimtes Cost Benefit Anlyses Technicl evlutions Conclusions nd limittions Pge 2

3 Introduction Generl Sfety Regultion ims to mndte AEBS nd LDWS for ll N2, N3, M2 nd M3 vehicles; Exemptions permitted if cost/benefit nlysis nd/or other vehicle-specific sfety spects show fitment to be inpproprite; TRL tsked by EC to follow up its erlier nlyses on N2/N3 nd M2/M3 with more detiled nlyses of specific vehicle types. Use of GB, Germn nd French dt to estimte EU27 trget popultions, i.e. ll csulties tht might be preventble by 100% effective AEBS or LDWS; Considertion of effectiveness, system costs nd vehicle stock/new registrtion dt to produce vehicle-specific costbenefit nlyses. Desk-bsed ssessment of suggested technicl exemptions. Pge 3

4 Vehicle Types N2/N3 vehicle types tht re possible to identify or pproximte, e.g. by body type, GVW. Most detiled vehicle type cn be pplied to ech of the vehicle types bove it, e.g. Off rod vehicles cn be N2, N3 16t or N3>16t ll vilble combintions re shown in the lter slides. Vehicle Type Comment Goods vehicles >3.5t nd lrge vehicles clssed s other N2/N3 motor vehicles such s refuse collectors, mobile crnes, recovery vehicles N2 GVW<=12t N3 GVW>12t N3 16t GVW >12t nd <=16t N3 >16t GVW >16t Off rod Rigid vehicles only where body types re tipper/concrete mixer Specil purpose Rigid vehicles with body types such s brekdown truck, skip loder, street clenser, rod stripper etc. >5t nd 7.5t N2 restricted by GVW 5t Pge 4 N2 restricted by GVW > 7.5t N2 restricted by GVW Trctor 7.5t Vehicles with n rticulted wheel-pln, restricted by GVW >3 xles Not intended to tow O3 or O4 triler Intended to tow O3 or O4 triler with more thn 3 xles nd triler for exceptionl lod trnsport nd trilers with res for stnding pssengers Rigid vehicles with more thn 3 xles or rticulted vehicles with tow vehicle with more thn 3 xles Cnnot identify vehicles not designed to tow, but cn identify rigid vehicles tht were not towing t the time of the ccident likely to result in over-estimte Not possible to identify specific trilers, but possible to identify vehicles with GVW>44T which re likely to include some of these vehicles

5 Vehicle Types M2/M3 vehicle types tht re possible to identify or pproximte, e.g. by seting cpcity Vehicle Type Comment M2/M3 M2 M3 M3 Clss A M3 Clss I, II & III (often, occsionlly or no stnding pssengers) M3 >3 xles M3 rticulted Vehicles designed to crry more thn 8 pssengers in ddition to the driver Pssenger vehicles with 8 to 16 pssenger sets in ddition to the driver (e.g Ford Trnsit, Mercedes Sprinter). GVW for pssenger vehicles is not s robust s for goods vehicles Could over-estimte becuse could possibly include buses with 16 sets nd room for stnding pssengers, lthough subject to interprettion of the vehicle type by the police officer ttending the scene (not necessrily trffic officer) Pssenger vehicles with more thn 16 sets in ddition to the driver Vehicles with more thn 16 nd less thn 23 pssenger sets, includes ssumption tht minibuses re less thn 5T (M2) Not possible to differentite between buses with seted/stnding pssengers. Could be pproximted by considering ccident loction (urbn/rurl) Rigid vehicles with more thn 3 xles or rticulted vehicles with tow vehicle with more thn 3 xles Vehicles with >16 pssenger sets tht re rticulted Pge 5

6 Identifiction of vehicles types in ccident dt sources Vehicle Type GB DE FR All vehicle types N2/N3 N2 N2 Off rod N2 Specil purpose N2 > 3 xles N2 <5T N2 >5T <7.5T N2 > 7.5T N2 Trctor <7.5T N3 N3 <=16t N3 <=16t off rod N3 <=16t specil purpose N3 <=16t >3xles N3 <=16t rigid (not towing t time) N3 >16t rigid (not towing t time) Pge 6 N3 >44t exceptionl lod N3 >16t N3 >16t off rod N3>16t specil purpose N3 >16t >3xles Dt Sources Gret Britin Ntionl rod ccident sttistics (Stts19), nd Stts19 linked to Vehicle Registrtion dt Germny Gids dt from weighted to Ntionl Sttistics for 2008 Frnce NtionlN l ccident sttistics ti ti 2007 Vehicle Type GB DE FR M2/M3 M2 M3 M3<= 23 sets M3>23 sets M3>3 xles M3 rticulted Pge 6

7 Trget Popultions - LDWS Hed-on Collisions M2/M3/N2/N3 unintentionlly leving the lne Leving Crrigewy M2/M3/N2/N3 unintentionlly leving the lne Side-swipe M2/M3/N2/N3 unintentionlly leving the lne All LDWS relevnt ccidents All ccident loctions Pge 7 The unintentionl lne chnge of the vehicle of interest is identified bsed on number of criteri: the mnoeuvre being crried out (generl driving, not mking specific mnoeuvre) The The impct loction on the vehicle of interest nd the other vehicle Built-up/non built-up/ motorwy Driver behviour fctors tht could contribute to leving the lne unintentionlly e.g., ftigue, distrction, etc

8 Trget Popultions - AEBS Dt filtering Front of M2/M3/N2/N3 to rer of other vehicle Non- vulnerble rod user csulties ( ll rer shunts ) Resulting in csulties tht re vulnerble rod users Sttionry trget vehicle Non-sttionry trget vehicle Groups of interest in Ornge; Current sytems require the trget vehicle to be trcked while it is moving. It is not possible to differentite between vehicles moving or not moving prior to impct; The closest pproximtion is to identify trget vehicles tht re likely to be sttionry on impct by selecting mnoeuvres such s prked, held up, witing to turn; Hve lso ssessed ll rer shunt ccidents, to quntify potentil of future systems tht cn recognise sttionry nd non-sttionry trget vehicles. Pge 8

9 EU Accident/Csulty Sttistics EU27 DG-TREN Sttisticl Pocketbook 2010 Mesure Accidents ll severities Annul Averge 1,321,450 1,299,245 1,296,928 1,232,211 1,287,459 Ftlities 45,300 43,062 42,496 38,875 42,433 EU16 from CARE where vehicle types, ll severities nd yers re vilble nnul verge Pge 9 Accident type Annul verge number of csulties (EU16*) Ftl Serious Slight All 28, ,450 1,051,960 Involving N2/N3** 4,803 16,597 63,669 Involving M2/M ,945 33,602 Austri, Belgium, Czech Republic, Germny, Denmrk, Spin, Frnce, United Kingdom, Greece, Hungry, Mlt, Netherlnds, Portugl, Romni, Sweden, Sloveni, ** Dt for other motor vehicles (OMV) only vilble for for 14 of 16 Member Sttes. OMVs for GB included griculturl nd hve been updted using Stts19 dt to remove the griculturl vehicles

10 EU27 Csulty Estimtes Overll number of ftlities known for EU27, so proportions involved in N2/N3 or M2/M3 in EU16 ssumed to pply to EU27, e.g. M2/M3 ftlities = (42433 x 822/28828) = 1210; For non-ftl csulties, overll number of EU27 ccidents ssumed to be split by severity (ftl, serious nd slight) in sme proportions s EU16, nd numbers of csulties lso ssumed to be in sme rtios per ccident for ech severity; These estimtes, though, re bsed only on reported ccidents, so represent lower estimte. Mid estimte mkes some llownce for under-reporting by using GB & Germn proportions of ftl:serious:slight (ftlities sme s lower estimte). Upper estimte goes one step further nd uses best vilble reserch to mke llownce for under-reporting in GB nd Germny, nd pply smll correction for deths > 30 dys. Pge 10

11 EU27 Csulty Estimtes Estimtes for ll N2/N3 nd M2/M3 csulties in EU27. Vlutions bsed on 2008 UK costs. Accident type Estimted nnul verge number of Vlue csulties (EU27) Ftl Serious Slight All m Lower estimtes Involving N2/N3 (rtios) 7,070 (1:) 21,950 (3:) 84,203 (12) 113,223 21,611 Involving M2/M3 (rtios) 1,210 (1:) 6,540 (5:) 44,439 (37) 52,189 4,905 Mid estimtes Involving N2/N3 (rtios) 7,070 (1:) 35,352 (5:) 212,109 (30) 254,531 27,115 Involving M2/M3 (rtios) 1,210 (1:) 12,102 (10:) 102,869 (85) 116,181 7,287 Upper estimtes Involving N2/N3 (rtios) 7,212 (1:) 44,189 (6:) 371,191 (51) 422,592 32,406 Involving M2/M3 (rtios) 1,234 (1:) 15,128 (12:) 180,021 (146) 196,383 9,462 Pge 11

12 Vehicle stock nd new registrtion estimtes No comprehensive published dt on numbers of vehicles in use cross EU27 by vehicle types of interest; For N2/N3, EU27 stocks hve been estimted bsed on pproprite combintions of EU27 dt for ll goods vehicles in use nd for new vehicle registrtions (for N2/N3 vehicles nd N3 > 16t), from DG-TREN Pocketbook, ntionl dt from GB nd Germny, nd dt provided by industry stkeholders; For M2/M3, EU27 stocks nd new registrtions published in Pocketbook. Specific vehicle types estimted using GB nd Germny dt. Estimtes of 5.7million N2/N3 vehicles in use (2.5m N2, 3.2m N3), nd 883,000 M2/M3 (221,000 M2, 662,000 M3). Estimtes of 423,000 N2/N3 nd 57,000 M2/M3 new registrtions per yer (bsed on verges). Pge 12

13 System costs nd effectiveness estimtes Reserch literture review nd figures provided by stkeholders (for erlier projects) used to mke estimtes: AEBS (ll rer shunts): , 20-50% effectiveness; AEBS (non sttionry trgets): , 20-50% effectiveness; LDWS: , % effectiveness. Pge 13

14 Cost Benefit Anlyses First stge: simple brek-even nlysis bsed on trget popultions nd new registrtions only; Second stge: Phsed nlysis llowing for effectiveness nd grdul introduction into vehicle fleet over 15 yer evlution period (vi mndtory fitment to ll new vehicles). Results in form of brek-even costs (mximum cost of systems per vehicle to provide benefit-cost rtio > 1) nd benefit:cost rtios (bsed on estimted system cost rnges). Pge 14

15 Results - AEBS (ll rer shunts) Vehicle Type Stock (1000s) New registrtions per nnum (1000s) EU-27 nnul estimtes Trget popultion (Csulties) Simple Brek-Even Costs ( ) (if systems 100% effective) AEBS (ll rer shunts) Phsed Brek-Even Costs ( ) (15 Yrs, fit to ll new vehicles) Phsed Benefit:Cost Rtios (15 Yrs, fit to ll new vehicles) Lower Mid Upper Lower Mid Upper Lower Mid Upper Lower Mid Upper Effectiveness 20% 35% 50% System costs 1, N2/N ,349 7,247 9, ,482 2, N ,964 7,815 9, ,908 3, N3 >16t ,600 7,337 9, ,821 3, N ,507 5,546 7, , N2 >5 <=7.5t ,480 5,503 7, , N3 >16t >3 xles ,282 8,232 10, ,173 3, N3 >16t Off Rod ,364 8,339 10, ,988 3, N3 <=16t ,372 22,762 28,519 1,329 3,047 5, N3 <=16t Rigid (not towing) ,473 21,584 27,044 1,244 2,852 5, N2 >7.5t ,072 9,602 13, , N3 >16t Rigid (not towing) N2 <=5t ,204 3,485 4, , N3 >16t Specil Purpose ,333 3,056 3, , N2 Specil Purpose ,962 4,684 6, , N2 Off Rod ,602 2,534 3, N3 <=16t Specil Purpose ,432 30,703 38,468 2,451 5,621 10, N3 >44t?? ????????? N3 <=16t Off Rod ,493 26,664 33, ,562 2, N2 >3 xles * ,267 5, , N3 <=16t >3 xles * ,522 49, ,145 5, N2 Trctor unit <=7.5t * M2/M ,059 6,615 9, ,196 2, M ,961 8,086 11, ,462 2, M3 Clss I/II/III ,478 8,928 12, ,608 3, M ,351 2,202 3, M3 Clss A ,287 1, M3 rticulted * ,512 3, , M3 >3 xles * ,798 6,058 8, ,784 3, * NB. The numbers of these vehicles in use in GB re very low (dt from other countries not vilble), so very low numbers of csulties re to be expected - EU-27 estimtes re thus subject to significnt uncertinty. Pge 15

16 Results - AEBS (non sttionry trgets only) Vehicle Type Stock (1000s) New registrtions per nnum (1000s) EU-27 nnul estimtes AEBS (non sttionry trgets) Trget popultion (Csulties) Simple Brek-Even Costs ( ) Phsed Brek-Even Costs ( ) (if systems 100% effective) (15 Yrs, fit to ll new vehicles) Phsed Benefit:Cost Rtios (15 Yrs, fit to ll new vehicles) Lower Mid Upper Lower Mid Upper Lower Mid Upper Lower Mid Upper Effectiveness 20% 35% 50% System costs 1, N2/N ,718 3,691 4, , N ,901 3,898 4, , N3 >16t ,845 3,823 4, , N ,177 3,081 3, N2 >5 <=7.5t ,327 3,292 4, N3 >16t >3 xles ,965 3,984 5, ,052 1, N3 >16t Off Rod ,589 3,479 4, , N3 <=16t ,642 6,238 7, , N3 <=16t Rigid (not towing) ,167 5,600 7, , N3 >16t Rigid (not towing) N3 >16t Specil Purpose ,035 1,391 1, N2 >7.5t ,021 4,274 5, N2 <=5t ,242 1,758 2, N2 Specil Purpose ,030 2,873 3, N2 Off Rod ,394 1, N3 <=16t Specil Purpose ,470 10,039 12, ,838 3, N3 >44t?? ????????? N3 <=16t Off Rod ,558 10,393 13, , N2 >3 xles * ,589 4, , N3 <=16t >3 xles * ,405 19, ,226 2, N2 Trctor unit <=7.5t * M2/M ,400 3,273 4, , M ,836 3,867 4, , M3 Clss I/II/III ,139 4,280 5, , M ,093 1,491 1, M3 Clss A M3 rticulted * ,544 1,760 2, M3 >3 xles * ,661 3, , * NB. The numbers of these vehicles in use in GB re very low (dt from other countries not vilble), so very low numbers of csulties re to be expected - EU-27 estimtes re thus subject to significnt uncertinty. Pge 16

17 Results - LDWS Vehicle Type Stock (1000s) New registrtions per nnum (1000s) EU-27 nnul estimtes LDWS Trget popultion (Csulties) Simple Brek-Even Costs ( ) Phsed Brek-Even Costs ( ) (if systems 100% effective) (15 Yrs, fit to ll new vehicles) Phsed Benefit:Cost Rtios (15 Yrs, fit to ll new vehicles) Lower Mid Upper Lower Mid Upper Lower Mid Upper Lower Mid Upper Effectiveness 20% 40% 60% System costs N2/N ,830 4,853 5, ,134 2, N ,615 4,623 5, ,290 2, N3 >16t ,574 4,571 5, ,296 2, N ,474 5,537 6, , N2 >5 <=7.5t ,194 5,191 6, , N3 >16t Rigid (not towing) ,296 1,658 1, N2 >7.5t ,746 10,824 12, ,230 2, N3 >16t Off Rod ,685 2,155 2, , N2 <=5t ,080 3,812 4, , N2 Off Rod ,752 5,881 6, , N3 >16t >3 xles ,248 1,596 1, N3 <=16t ,881 6,243 7, , N3 <=16t Rigid (not towing) ,712 6,026 7, , N3 >16t Specil Purpose ,307 1,672 2, N2 Specil Purpose ,580 3,193 3, , N3 >44t?? ????????? N3 <=16t Specil Purpose ,780 12,508 14,978 1,023 2,617 4, N3 <=16t Off Rod ,230 28,474 34, ,907 3, N2 >3 xles * ,057 18,676 21,894 2,141 4,980 8, N3 <=16t >3 xles * ,205 50, ,838 6, N2 Trctor unit <=7.5t * , M2/M ,356 11,501 13, ,376 4, M ,768 14,465 16,849 1,215 2,988 5, M3 Clss I/II/III ,992 15,971 18,603 1,338 3,288 5, M ,122 2,608 3, M3 rticulted * ,558 37,722 43, ,877 17, M3 Clss A ,875 2,304 2, M3 >3 xles * , , ,619 53, * NB. The numbers of these vehicles in use in GB re very low (dt from other countries not vilble), so very low numbers of csulties re to be expected - EU-27 estimtes re thus subject to significnt uncertinty. Pge 17

18 Results - observtions For AEBS, only N2 trctor units <= 7.5t give benefit-cost rtios < 1 in ll three scenrios; For LDWS, no vehicle types hve rtio < 1 in ll scenrios; Assuming low trget popultions, low effectiveness nd high costs gives benefit-cost rtios very often < 1; In the mid scenrio, rtios re usully well bove 1 for LDWS nd close to 1 (+/-) for AEBS; Results for N2/N3 nd M2/M3 in line with erlier reserch; Hevier vehicles (N3 nd M3) tend to give higher benefit-cost rtios thn the lighter ones (N2 nd M2); Results generlly more sensitive to rnges of cost estimtes (e.g. 1:6) thn effectiveness (1:3) or csulty vlutions (1:2) Pge 18

19 Technicl Evlutions Vrious stkeholders hve provided technicl justifictions for the vehicles they propose should be exempt. TRL hve reviewed these bsed on literture, ccident dt nd further input/explntion from stkeholders; For AEBS, min view tht would be prohibitively expensive for vehicles not equipped with EVSC nd tht, therefore, the sme exemptions should pply to AEBS s lredy pply to EVSC; Full evlution not possible becuse no evidence on the extr costs provided. Resonble, however, to postulte tht systems could be developed tht do not rely on the presence of EVSC. Costs would depend on functionlity nd the numbers of vehicles over which the development costs could be spred; An lterntive pproch my be to require forwrd collision wrning systems to lert the driver to pply the brkes if full AEBS is not fesible or cost effective. For LDWS, the stkeholder view tht systems designed only to operte t speeds over 60 km/h would be of little use to vehicles used minly off-rod or in urbn res is prtly supported by the ccident dt, but not enough to suggest tht such systems could not still provide benefit-cost rtio of more thn 1. Pge 19

20 Conclusions & Limittions Decisions regrding the pplicbility of the cost-benefit nlysis nd technicl evlutions to questions of possible exemption for specific vehicle types re politicl mtter for stkeholders nd legisltors to debte nd decide upon; Importnt, though, to remember tht the technicl ssessments, benefit-cost rtios nd brek-even costs presented in this report re ll bsed on wide vriety of ssumptions nd subject to limittions, e.g: - Pst ccident/csulty sttistics re imperfect predictors of future ptterns. - Accident nlyses bsed on smples of dt from, t best, 3 Member Sttest nd often only 1 or 2, giving high level of uncertinty when mking EU27 estimtes, prticulr where low numbers of vehicles nd ccidents combine. - The vehicle types ssessed within the ccident dtbses re, generlly speking, not exct mtches to the vehicle types proposed by stkeholders for exemptions. - Different AEBS nd LDWS rchitectures will hve different opertionl chrcteristics, costs nd effectiveness in differing pplictions nd ccident scenrios. The nlyses presented here re entirely generic nd cn only ttempt to llow for these vritions by using quite wide rnges of likely effectiveness nd costs. Pge 20

21 THANK YOU! Brin Robinson, Iin Knight, Pge 21