SPEED CONTROL EVALUATION ON THE A13 MOTORWAY (FRANCE)

Transcription

1 EasyWay Evaluation Expert Group (EW EEG) Document EEG/11/7 SPEED CONTROL EVALUATION ON THE A13 MOTORWAY (FRANCE) Final Version Issue: 2 nd edition Version: 0.2 Date: 13 th September 2012

2 TABLE OF CONTENTS REPORT TEMPLATE 5 Project Reference (From EasyWay Work plan): 5 Project Name: Régulation des vitesses sur A13 (Variable speed limit on the A13 motorway) 5 EasyWay Region: Centrico 5 1. Key Evaluation Results Impact on Traffic Flow WEEK DAYS EVALUATION WEEK-END EVALUATION Impact on Safety Impact on Environment Other Key Results WORKING DAYS EVALUATION 9 2. Description of the Problem Site Issues Addressed Description of the ITS Project Service Area Key Words Objectives 11 13/09/ Speed Control Evaluation on the A13 Motorway (France)

3 3.4. Systems and Technologies Applied Costs Status of the Project Evaluation Planned Timing and Type of Evaluation Objectives for the Evaluation Research Questions Study Area for the Evaluation Expected Impacts Expected Methods The Impact of the Project - Results Technical Performance Results RESULTS OF VSL FOR WEEK DAYS RESULTS OF VSL FOR WEEK-END Reliability of Results RAW TRAFFIC DATA QUALIFICATION SIMILARITY OF SAMPLE SETS USED FOR COMPARISON CROSS VALIDATION Research Questions Answered Overall Assessment SAFETY EFFICIENCY ENVIRONMENT 31 13/09/ Speed Control Evaluation on the A13 Motorway (France)

4 6. European Dimension: Transferability of the Results 32 13/09/ Speed Control Evaluation on the A13 Motorway (France)

5 REPORT TEMPLATE Project Reference (From EasyWay Work plan): Project Name: Régulation des vitesses sur A13 (Variable speed limit on the A13 motorway) EasyWay Region: Centrico 13/09/ Speed Control Evaluation on the A13 Motorway (France)

6 1. Key Evaluation Results This section should present a brief summary of the key results related to the EasyWay objectives and any other key results. The evaluation of the A13 variable speed limit system has been divided into two parts. First part is the evaluation of variable speed limit over working days, and the second part is the evaluation of variable speed limit during week-end. Therefore, 2 sets of results will be given here below, one for each evaluation part. The sections below show the key evaluation results. Detailed results are provided in chapter Impact on Traffic Flow WEEK DAYS EVALUATION CONGESTION INTENSITY During week days congestion have decreased by 50%. The graphic below depicts the total length of congestion per month before (light blue) and after (dark blue) the variable speed limit was started. We can see that the variable speed limit has brought a great improvement in traffic flow during peak traffic periods. Comparaison des HKM Avant - Après RDV JUIN SEPTEMBRE OCTOBRE HKM AVANT RDV HKM APRES RDV LEVEL OF SERVICES The level of service represents the user perception of its journey through the motorway (see explanation of this indicator in chapter 4.6). A good level of service would be when the traffic flow is fluid. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

7 The analysis of the level of services shows that these levels have increased. It can directly be seen on the graphics below where the green colour represents the best level of services while the black is a bad level of service. September and October 2008 and 2009 are compared for 2 measurement station. Station S02 - Septembre - Comparaison des niveaux de service entre 2008 et 2009 Station S02 - Octobre - Comparaison des Niveaux de service entre 2008 et % 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Niveaux de service 2008 V1 = 111 km/h V2 = 106 km/h V3 = 68 km/h % 80% 70% 60% 50% 40% 30% 20% 10% 0% Niveaux de service 2008 V1 = 110 km/h V2 = 102 km/h V3 = 60 km/h Station IR Septembre - Comparaison des Niveaux de service entre 2008 et 2009 Station IR102 - Octobre - Comparaison des niveaux de service entre 2008 et % 90% 80% 70% 60% 50% 40% 30% Niveaux de service 2008 V1 = 107 km/h V2 = 100 km/h V3 = 59 km/h 100% 90% 80% 70% 60% 50% 40% 30% Niveaux de service 2008 V1 = 107 km/h V2 = 100 km/h V3 = 58 km/h 20% 20% 10% 10% 0% % WEEK-END EVALUATION CONGESTION INTENSITY The comparison between the number of congestions before and after, do not show an improvement after variable speed limit was set on for week-end days. On the contrary, the graphic below show harder congestion after than before. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

8 HKM Comparaison des HKM Avant RDV / Après RDV selon le type de retour de week end (Trafic faible / trafic fort) HKM Avant RDV HKM Après RDV Trafic Faible Trafic Fort LEVEL OF SERVICES Correlatively to what was seen on the congestion intensity, the levels of services do not show any improvement during week-end with variable speed limit on. The figures below show that the levels of services are even worst while variable speed limit is activated. Niveau de service - Station S05 - trafic fort Niveau de service - Station S03 - trafic fort 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Avant RDV Après RDV 100% 80% 60% 40% 20% 0% Avant RDV Après RDV Note that «Avant RDV» stands for Before VSL and Après RDV stands for With VSL 1.2. Impact on Safety The table below shows the variation in the number of accidents on the past 3 years on the road section with traffic control. It shows a decrease in the number of accidents in 2009, year where the variable speed limit system was set on. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

9 1.3. Impact on Environment This impact was not measured Other Key Results A summary of the key results for working days is presented here below WORKING DAYS EVALUATION 13/09/ Speed Control Evaluation on the A13 Motorway (France)

10 2. Description of the Problem 2.1. Site The site is located in France, on the A13 motorway, West of Paris. It is a 25km long section of motorway where daily traffic reaches vehicles. See map below. Niveau Issues Addressed During morning peaks, 6000 vehicles per hour travel in direction of Paris often causing strong congestion and accidents. Therefore, in order to improve traffic flow on this section it was decided to implement a variable speed limit system. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

11 3. Description of the ITS Project The ITS project consists in several topics: The development of a variable speed limit algorithm The development of the entire chain from field data measurements to VMS indicating updated speed limits on the motorway Field deployment of several VMS for speed display End user information 3.1. Service Area The EasyWay Service areas are: Traffic Management Services : deployment of variable speed limits 3.2. Key Words Variable speed limits, speed control, dynamic speeds, VMS, flow, capacity, traffic jam, congestion, travel time, traffic control centre, traffic indicators, accordion driving 3.3. Objectives The project objectives were to: decrease congestion intensity improve traffic flow improve safety reduce impact on environment Being an experiment, initial project objectives were not quantified Systems and Technologies Applied The picture below fully describes the ITS project : 13/09/ Speed Control Evaluation on the A13 Motorway (France)

12 The ITS project was divided in several work packages, as follow: Variable speed limit algorithm o Traffic data study o VSL algorithm definition o VSL algorithm simulation o VSL algorithm calibration Variable speed limit software o Software specification o Software provider selection o Software development o Software validation o Software deployment VMS definition o VMS location study o Specification for VMS o VMS provider selection o VMS development o VMS validation o VMS installation Traffic Control Center o VSL installation at TCC o User information and training o Maintenance team training Customer information o Fixed signs along motorway o Communication plan on the motorway FM radio Administrative task o Informing authorities about the project o Obtaining authorisation from state and local authorities 3.5. Costs Capital costs - ITS Technologies o Studies : 50 k o Software development : 220 k Capital costs Infrastructure o 200 k (counting station are not included as they were already in place) Installation costs Technologies and Infrastructure o 330 k Annual operating costs o Included in maintenance costs 13/09/ Speed Control Evaluation on the A13 Motorway (France)

13 Annualised maintenance costs o 12 k 3.6. Status of the Project Project is deployed since April VSL has proved to be very efficient during working days peak traffic in the morning. We are currently working on parameters adjustments for week-end as VSL has currently not proved to be efficient during week-end return to Paris. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

14 4. Evaluation Planned 4.1. Timing and Type of Evaluation Please state whether the evaluation is carried out prior to implementation (ex-ante evaluation), or following implementation (expost evaluation), and whether or not simulation is involved. The evaluation was carried out following implementation (ex-post evaluation). Two evaluations were performed: one for week days one for week-end days The first evaluation was build using a set of working days from June, September and October The second evaluation was made using a set of week-ends from 2009 and Objectives for the Evaluation The VSL is a new system. Therefore, an evaluation is necessary to assess the capacity for the VSL system to improve traffic conditions, safety and the environmental impact. Depending on the evaluation results, the system can be deployed in other places and will serve as a reference model for other road authorities willing to deploy such systems on peri-urban motorways. The main objectives pursued in evaluating the VSL system are: Traffic conditions : o Verify that the infrastructure is used optimally. o Verify that the flow has increased o Verify journey time evolution o Measure impact on traffic jam and congestion intensity o Measure impact on the level of services Safety o Ensuring that speed limits indicated by VMS are respected; o Check the evolution of the number of accidents Environment: o Verify the impact on environment 4.3. Research Questions Describe the research questions posed by the various stakeholders involved before the evaluation started (e.g. government, researchers, public), and which of these were covered in the evaluation. The VSL experiment was authorized and followed by the French Ministry of Transport (Ministère de l Ecologie, du Développement durable, des Transport et du Logement / Direction générale des Infrastructures, des Transports et de la Mer). Behind the experiment, several questions were asked; in particular, qualify the ability of the VSL system to improve traffic indicators, 13/09/ Speed Control Evaluation on the A13 Motorway (France)

15 4.4. Study Area for the Evaluation Describe the study area and show how the area covered by data collection relates to the ITS site described in Section 2.1 (e.g. does the evaluation cover wider impacts on the network or is it confined to the immediate site of the ITS scheme). The study area is the section where the VSL system has been implemented. This is the motorway section between Mantes La Jolie and Orgeval, on the A13 motorway, France, west of Paris Expected Impacts Describe the expected impacts, and identify those that will be assessed in the evaluation. As a minimum these should relate to those EasyWay objectives addressed by the project (efficiency, safety, environment related) The main impacts expected by the system, are: Efficiency o Homogenization of speed and reduction of instability phenomena, with consequent higher fluidity of the traffic flow. o Lower travel times o Improve infrastructure capacity o Reduction of the time under congestion. Safety: o Reduction of stop & go o Better level of service for the end user o Harmonization of speeds o Reduction in the number of accidents Environment o Reduction of the emissions of polluting agents 4.6. Expected Methods Describe the methods to be used to inform each of the objectives for the evaluation (identified above) under the following headings. The following indicators have been used for the days of week evaluation: Number of days where speed were changed Mean speeds: this indicator measures the mean speeds at several stations on the section. The speeds are measured using either inductive loops stations or infrared sensors stations. Trafic Jam: this indicator measures the congestion length expressed in HKM, that is hour of congestions multiplied by length of congestion. The congestions lengths are manual estimation through video cameras. Travel times: this indicator measure the travel time on the section using 2 methods: one using toll tags information between 2 tolls plazas, the other using travel time calculated through the use of speeds measured by inductive loops. Lane capacity : this indicator measures the lane capacity using inductive loop station Level of services: this indicator is based on the infrastructure capacity and measure the user experience of its journey through the motorway section. A detailed description of how this indicator is calculated is provided in the figure below. Using this indicator, one can draw space/time diagrams of the level of services, making before and after data easily comparable. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

16 Number of accidents : this indicator is directly extracted from the traffic control centre information database Speed respect and vehicle inter-distances: these indicators are individual measurement of speeds and times between vehicles. They are measured using the individual information from inductive loops stations. The following indicators have been used for the week -end evaluation: Number of days where speed were changed Mean speeds: this indicator measures the mean speeds at several stations on the section. The speeds are measured using either inductive loops stations or infrared sensors stations. Using these values on each station one can draw space/time diagram of speeds which can be compared before and after. Trafic Jam: this indicator measures the congestion length expressed in HKM, that is hour of congestions multiplied by length of congestion. The congestions lengths are manual estimation through video cameras. Travel times: this indicator measure the travel time on the section using 2 methods: one using toll tags information between 2 tolls plazas, the other using travel time calculated through the use of speeds measured by inductive loops. Lane capacity : this indicator measures the lane capacity using inductive loop station Level of services: this indicator is based on the infrastructure capacity and measure the user experience of its journey through the motorway section. Using this indicator, one can draw space/time diagrams of the level of services, making before and after data easily comparable. Approach The approach used for the evaluation is a comparison before / after. All data were measured before the VSL system was started and compared to the same data with the VSL system turned on. To obtain sufficient sample of data to do the comparison, and also because traffic behaviours are very different, the study was divided in 2 parts: 13/09/ Speed Control Evaluation on the A13 Motorway (France)

17 uvp horaires Vitesses moyennes one for week days one for week-end days The comparison of before and after data needed to be done in similar conditions of traffic. The data with the VSL turned on are those of June / September and October months of year The before data to be used for the comparison were selected among the same months from 2007 or For the week days analysis, the selection of the best month for the before and after comparison, was made using a comparison of traffic flow at the entrance of the motorway section with VSL. This comparison allowed us to make sure that the levels of traffic were similar before and after the VSL system was turned on. As an example, the picture below show the comparison of traffic flows for the month of October, for which October 2008 (before data) was selected for the comparison with October 2009 (after data). So as a result we chose to compare the following month : June 2009 with June 2007 September 2009 with September 2008 October 2009 with October 2008 A13 - Débits horaires TV et vitesses moyennes horaires au poste de comptage S05 - sens 2 des JO moyens - Comparaison octobre 2007 à Débit moyen 2009 Débit moyen Heures 2008 Débit moy 2007 Vitesse moy 2009 Vitesse moy 2008 Vitesse moy For the week-end analysis, the comparison was done between Sundays afternoon before and after. In order to compare similar type of days, the week-ends were split into 2 categories: over and less than vehicles/day. In result, we selected 126 week-end for the before and after comparison : 31 week-ends in 2010 (from 05/05/10 to 31/10/10) 31 in 2009 (from 05/07 to 25/10) 31 in 2008 (from 06/04 to 26/10) 33 in 2007 (from 01/04 to 28/10) 13/09/ Speed Control Evaluation on the A13 Motorway (France)

18 Expected performance (definition of success) If we look at the indicators the definition of success for each of them would be as follow : Mean speeds: increase Trafic Jam: decrease Travel times: decrease Lane capacity : increase Level of services: more and levels, less and Number of accidents : decrease Speed respect : the less over passing vehicles the better Vehicle inter-distances: the longer inter-distance the better Data collection (include technical details such as questionnaires, model calibration etc in annex) For mean speeds, travel times, level of services The collected data consist mainly in the heavy/light vehicles flow, speed and occupancy measured by the mean of inductive loops or infrared sensors. They are collected with an aggregation of 6 minutes. For travel times, toll tag data were collected and travel time were extracted from these data Accidents were extracted from the traffic centre information database Speed respect and vehicle inter-distance were recorded on one counting station (station S2) The following picture shows the position of the counting stations on the section. Sx represents an inductive loop station IRx represents an infra-red counting station The VSL section is represented by the blue rectangle 13/09/ Speed Control Evaluation on the A13 Motorway (France)

19 uvp horaires Data selection (in cases where not all data collected were used in the analysis, handling of missing data or erroneous data etc) It was indeed necessary to remove not usable data from the data set. In order to detect days with potentially inappropriate data for the comparison, we draw for each station the flow diagram for each day. These flow diagrams were compared to a standard day diagram with bounding envelopes. The figure below describes show an example of the standard days with the bounding envelop in grey. A13 - Débits moyens horaires TV au poste de comptage S02 - sens 2 des JO moyens "standard" de septembre lundi moyen mardi moyen mercredi moyen jeudi moyen vendredi moyen fuseau - max fuseau - min heures Whenever the station diagram was not in the bounding envelope, we analysed the reasons of these discrepancies. It could have been due to days with an accident or specific days like bank holidays or first school day. In those cases and in cases where no apparent cause for the discrepancy were found, these days were removed from the sample set in order to retain the most homogenous sample set. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

20 5. The Impact of the Project - Results 5.1. Technical Performance Describe the technical performance of the system or service, and conclusions about the influence of technical performance on the impacts identified. The VSL system calculates on its own, through its internal algorithm, the speeds to be displayed on the VMS signs along the VSL section. Initially, the transmission of new speeds to the VMS had to be validated by the traffic control center operator whether speeds were lowered or increased. The operator had to verify the consistency of the proposed speed in regard of the current traffic situation. The VSL algorithm has proved to be so stable in its proposal for new speeds that it was decided to fully rely on its calculation and remove the operator validation. Since then, the system is fully automated and the system changes speeds up and down automatically depending on traffic flow. No human action is required. Another technical performance of this system is its ability in decreasing congestions during traffic peak of week days. The results below show this excellent capacity in reducing congestions, accidents, increasing mean speed and travel time Results This section is separated into 2 parts, one for week days, the second for week-end RESULTS OF VSL FOR WEEK DAYS MEAN SPEEDS COMPARISON 13/09/ Speed Control Evaluation on the A13 Motorway (France)

21 TRAFFIC JAM LEVEL OF SERVICES 13/09/ Speed Control Evaluation on the A13 Motorway (France)

22 TRAVEL TIME LANE CAPACITY 13/09/ Speed Control Evaluation on the A13 Motorway (France)

23 NUMBER OF ACCIDENTS SPEED RESPECT AND VEHICLE INTER-DISTANCE 13/09/ Speed Control Evaluation on the A13 Motorway (France)

24 RESULTS OF VSL FOR WEEK-END MEAN SPEEDS The study of mean speeds does not show any improvement in mean speeds. Moreover, it appears that mean speeds measured after VSL was started, are lower than before while we were expecting an improvement in mean speed as the number of congestion should have decreased. The tables below show the evolution of speeds before (see Avant RDV ) and after (see Après RDV ) VSL was started. The measuring stations are names Sxx or IRxxx. High Trafic Mean Speed Standard deviation Minimum Mean speed Flow S05 S03 S02 Avant RDV 98 km/h 16 km/h 78 km/h 5278 Après RDV 82 km/h 23 km/h 66 km/h 5169 % de variation - 19,0% + 27,7% - 18,1% - 2,1% Variation Avant RDV 95 km/h 19 km/h 74 km/h 4609 Après RDV 96 km/h 22 km/h 72 km/h 4602 % de variation + 1,3% + 12,4% - 2,6% - 0,1% Variation Avant RDV 86 km/h 24 km/h 62 km/h 4500 Après RDV 79 km/h 26 km/h 59 km/h 4507 % de variation - 8,4% + 7,2% - 4,9% + 0,2% 13/09/ Speed Control Evaluation on the A13 Motorway (France)

25 IR102 IR101 Variation Avant RDV 84 km/h 20 km/h 66 km/h 4098 Après RDV 78 km/h 30 km/h 58 km/h 4507 % de variation - 8,1% + 33,0% - 13,3% + 9,1% Variation Avant RDV 99 km/h 12 km/h 87 km/h 4356 Après RDV 93 km/h 13 km/h 84 km/h 4563 % de variation - 6,2% + 13,8% - 4,2% + 4,5% Variation Low Trafic Mean Speed*Standard deviation Minimum Mean speed Flox Avant RDV 104 km/h 12 km/h 89 km/h 5023 S05 Après RDV 95 km/h 18 km/h 71 km/h 5086 % de variation - 9,8% + 34,5% - 24,7% + 1,2% Variation Avant RDV 109 km/h 11 km/h 94 km/h 4230 S03 Après RDV 105 km/h 16 km/h 82 km/h 4297 % de variation - 3,5% + 33,4% - 15,9% + 1,6% Variation Avant RDV 99 km/h 17 km/h 74 km/h 4347 S02 Après RDV 91 km/h 21 km/h 64 km/h 4258 % de variation - 9,4% + 16,3% - 14,6% - 2,1% Variation Avant RDV 92 km/h 16 km/h 72 km/h 4233 IR102 Après RDV 90 km/h 23 km/h 69 km/h 4520 % de variation - 2,3% + 27,0% - 4,9% + 6,3% Variation Avant RDV 101 km/h 10 km/h 86 km/h 4366 Après RDV IR km/h 11 km/h 85 km/h 4385 % de variation - 6,3% + 6,2% - 1,2% + 0,4% Variation TRAFFIC JAMS The figure below shows the comparison of congestion before and after the VSL system was started for high and low traffic week-ends. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

26 13:00 13:24 13:48 14:12 14:36 15:00 15:24 15:48 16:12 16:36 17:00 17:24 17:48 18:12 18:36 19:00 19:24 19:48 20:12 20:36 21:00 21:24 21:48 22:12 22:36 23:00 23:24 23:48 00:12 00:36 01:00 01:24 01:48 HKM No improvement of traffic can be seen with the VSL. Comparaison des HKM Avant RDV / Après RDV selon le type de retour de week end (Trafic faible / trafic fort) HKM Avant RDV HKM Après RDV Trafic Faible Trafic Fort TRAVEL TIME For low and high traffic week-end, the travel time are longer with the VSL system turned on. Comparaison des temps de parcours entre Mantes et Poissy Retours de week-end à fort trafic TP Après RDV TP Avant RDV 13/09/ Speed Control Evaluation on the A13 Motorway (France)

27 13:00 13:24 13:48 14:12 14:36 15:00 15:24 15:48 16:12 16:36 17:00 17:24 17:48 18:12 18:36 19:00 19:24 19:48 20:12 20:36 21:00 21:24 21:48 22:12 22:36 23:00 23:24 23:48 00:12 00:36 01:00 01:24 01:48 Comparaison des temps de parcours entre Mantes et Poissy Retours de week-end à faible trafic TP moyen Après RDV TP moyen Avant RDV LEVEL OF SERVICES The level of services indicator does not show any improvement when the VSL system is turned on, being for high traffic week ends or low traffic week ends, as show in figure below. Low traffic week ends Niveau de service - Station S05 - trafic faible Niveau de service - Station S03 - trafic faible 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Avant RDV Après RDV 100% 80% 60% 40% 20% 0% Avant RDV Après RDV 13/09/ Speed Control Evaluation on the A13 Motorway (France)

28 Niveau de service - Station S02 - trafic faible Niveau de service - IR102 - trafic faible 100% 100% 80% 60% 40% 80% 60% 40% 20% 20% 0% Avant RDV Après RDV 0% Avant RDV Après RDV Niveau de service - IR101- trafic faible Station IR102 : no data 100% 80% 60% 40% 20% 0% Avant RDV Après RDV High traffic week ends Niveau de service - Station S05 - trafic fort Niveau de service - Station S03 - trafic fort 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Avant RDV Après RDV 100% 80% 60% 40% 20% 0% Avant RDV Après RDV 13/09/ Speed Control Evaluation on the A13 Motorway (France)

29 Niveau de service - Station S02 - trafic fort Niveau de service - IR102 - trafic fort 100% 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Avant RDV Après RDV 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Avant RDV Après RDV Niveau de service - IR101- trafic faible Station IR102 : dysfonctionnement de la station jusqu en juin 2008 inclus 100% 80% 60% 40% 20% 0% Avant RDV Après RDV 5.3. Reliability of Results RAW TRAFFIC DATA QUALIFICATION Results presented throughout the study are based on the analysis of traffic data logged in the company traffic database. All traffic data pass through several filters and qualification mechanisms which removes incoherent or wrong information from sensors. This data qualification system allows us to fully rely on traffic data when it has been qualified as valid. Therefore we can say that the traffic data used for this evaluation were reliable SIMILARITY OF SAMPLE SETS USED FOR COMPARISON Because the analyses method is based on a before and after comparison, we had to make sure that the sample set were comparable. That is why, this study started by making sure that the sample sets were similar in terms of traffic conditions, choosing the best before month for week-days comparisons or dividing weekends into 2 subsets based on traffic levels CROSS VALIDATION The results of this study were also compared with other informal feedback from other stakeholder like road users or traffic control centre operator. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

30 Regarding road users, the company s radio (107.7FM) made a survey, asking road users their feeling about the VSL system. The feedback was positive in all case. We also received some feedback through our websites, the majority of them were positive. The traffic control centre operators who have worked during the two conditions: before and after, really appreciate the VSL system as something helping in reducing the congestion level during morning peak rush. One of them even stating Since the VSL system has been activated, congestions have gone away 5.4. Research Questions Answered With reference to the research questions set out before the evaluation started (summarised in Section 0), describe the how the evaluation provided answers to the questions posed. If we refer to the first question: the ability of the VSL system to improve traffic indicators, we have seen in the study for week days that the system really has the capacity of : mitigating the congestions level decreasing travel times improving the level of services reducing the number of accidents For weekends, the VSL system does not appear to be efficient. We still have to analyse the reason of these deceiving results during the weekends Overall Assessment Summarise the main findings in terms of their impact on European objectives under the headings below SAFETY In terms of safety, the week days results have shown the system ability to reduce the number and the intensity of congestion. These good numbers have come along with a diminution of the number of accident. As speed and safety can be directly correlated, we can say that the speed limit system, by decreasing the speeds when traffic is high, does not only improve the flow but also improve safety by reducing the number of traffic events especially accidents EFFICIENCY The VSL system has proved to be really efficient during week days. It did not appear to bring any positive impact during week end. Nevertheless we are still monitoring this situation and working on to improve its efficiency during week end. For that, we will adapt the system s algorithm thresholds and we will try to understand the driver s behaviour so that we can adapt the system to this specific kind of traffic that we encounter during weekends. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

31 ENVIRONMENT Even though we did not measure the environmental impact of the speed limit system, we know that limiting speeds, improving traffic flow, limiting the number of stop and go, and decreasing congestion will definitely reduce gas emission and therefore have a positive impact on environment. 13/09/ Speed Control Evaluation on the A13 Motorway (France)

32 6. European Dimension: Transferability of the Results Authors are asked to indicate the extent to which the results of the evaluation are applicable to other similar implementations in the same or different countries. This should include any special experimental, external or local conditions which may have influenced the results. This section is to enable the reader in judging the transferability of the results obtained. Where several evaluation reports have been produced in respect of similar implementations, this section of the report could be used to present comparative results and to draw out conclusions relating to the transferability of those results. Links to related evaluation reports might also be added in this section. The speed control system that is installed on the A13 motorway made possible to validate the theory that indicates that by reducing the speed, the flow can be increased. The positive results were definitely confirmed during week days for commuter s traffic. Nevertheless the system did not improve the flow capacity during week-end days. This can be explained by the difference of drivers behaviours between week days and week-ends. So regarding the applicability of this system to other places, we would think that, if used on peri-urban motorways during commuters traffic peak, the system would be efficient. 13/09/ Speed Control Evaluation on the A13 Motorway (France)