brt workshop - bus rapid transit and traffic safety
Post on 19-Jan-2015
Embed Size (px)
DESCRIPTIONO Centro de Excelência em BRT Across Latitudes and Cultures (ALC-BRT CoE) promoveu o Bus Rapid Transit (BRT) Workshop: Experiences and Challenges (Workshop BRT: Experiências e Desafios) dia 12/07/2013, no Rio de Janeiro. O curso foi organizado pela EMBARQ Brasil, com patrocínio da Fetranspor e da VREF (Volvo Research and Education Foundations).
- 1. Bus Rapid Transit and Traffic Safety Luis Antonio Lindau, Director, EMBARQ Brasil Nicolae Duduta, Associate Transport Planner, EMBARQ July 12, 2013
2. The impact of BRT systems on traffic safety Factors that influence safety on BRT corridors Relationship between safety and operational performance Summary 3. Mexico CityGuadalajara Bogota Curitiba Porto Alegre Istanbul Delhi Ahmedabad Vancouver Brisbane Rio, SP, BH PereiraCali In Brazil it comprised: road safety audits and inspections in 5 cities 190 km of corridors more than 2 million pax per day A global study 4. The impact of BRT systems on traffic safety Factors that influence safety on BRT corridors Relationship between safety and operational performance Summary 5. Overall safety impact of a BRT Case study: Macrobs, Guadalajara Calz. Independencia, 2007 (before BRT implementation) 6. Overall safety impact of a BRT Case study: Macrobs, Guadalajara Reduction in the number of lanes Shorter pedestrian crossings Central median Existing buses and minibuses replaced with a single operating agency 7. Monthly crashes before and after the implementation of the BRT 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 0 50 100 150 200 250 Jan-07 Mar-07 May-07 Jul-07 Sep-07 Nov-07 Jan-08 Mar-08 May-08 Jul-08 Sep-08 Nov-08 Jan-09 Mar-09 May-09 Jul-09 Sep-09 Nov-09 Jan-10 Mar-10 May-10 Jul-10 Sep-10 Nov-10 Jan-11 Mar-11 May-11 Monthlycitywidecrashes(excludingtheBRTcorridor) MonthlycrashesontheBRTcorridor Citywide Crashes on the BRT Before During BRT construction During BRT 8. Impact on crashes by severity, per year Impact on traffic fatalities Annual fatalities in absence of BRT Annual fatalities with BRT Change in fatalities Baseline data Modeled baseline data Project data Modeled project data Macrobus BRT 1 3.5 0.3 1 - 2.5 Impact on traffic injuries Annual traffic injuries in absence of BRT Annual injuries with BRT Change in injuriesBaseline data Project data Macrobus BRT 96.8 30 - 66.8 Impact on all crashes Annual crashes in absence of BRT Annual crashes with BRT Change in crashesBaseline data Project data Macrobus BRT 2,341 1,010 - 1,331 9. 5000 3194 1 BRT lane 2 general traffic lanes Passenger per hour per direction (peak) 6 726 1 BRT lane 2 general traffic lanes Crashes per year Comparison between the bus lanes and the mixed traffic lanes 10. Overall safety impact of a BRT Case study: TransOeste, Rio de Janeiro Av. das Americas 2010 Av. das Americas 2012 11. Overall safety impact of a BRT Case study: TransOeste, Rio de Janeiro The layout of the BRT is very similar to Macrobus center lanes, median stations, overtaking lanes But the street is considerably wider (up to 100 m) and speed limits are as high as 80kmh 12. Impact on safety? There is not yet enough data available to evaluate TransOeste However, the data so far show between 1.5 and 4.5 fatalities per month on Av. das Americas in 2012, some of them involving BRT vehicles, which could mean an increase in fatalities after BRT implementation 13. BRTs have the potential to significantly improve safety on the streets where they are implemented BRT systems where reductions in injuries and fatalities have been observed include Macrobus (Guadalajara), Metrobus (Mexico City), TransMilenio (Bogota), Janmarg (Ahmedabad) Reductions range from 30% (Metrobus) to 70% (Macrobus) But some systems have seen increases in fatalities (Delhi) or have a concerning number of fatalities (TransOeste) Overview of safety impacts 14. The impact of BRT systems on traffic safety Factors that influence safety on BRT corridors Relationship between safety and operational performance Summary 15. Methodology Crash frequency models Road safety inspections Discussions with BRT agency safety and operations staff Factors that influence safety on BRT corridors 16. Statistical models that aim to explain the differences in crash frequencies (or crash rates) at different locations (e.g. intersections) using variables related to geometry, traffic volumes, land use, etc. Commonly use a Poisson or negative binomial (Poisson- Gamma) distribution Crash frequency models 17. Crash frequency models Variables (Xi) Coefficients (, i) P Annual average daily traffic (AADT, thousands of vehicles) 0.016 0.074 Total length of all approaches to the intersection (L, meters) 0.003 0.010 Average number of lanes per approach 0.334 0.000 Cross street is through street (=1 if yes, =0 otherwise) 1.142 0.029 Major T junction (=1 if yes, =0 otherwise) 0.719 0.019 Constant -3.914 0.000 N = 133, LR 2 (prob.) = 64.62 (0.000), Log likelihood = -141.580 Variables (Xi) Coefficients (, i) P Presence of a center median (=1 if yes, =0 otherwise) -0.349 0.004 Total number of approaches to the intersection (m) 0.424 0.000 Average length of approaches to the intersection (Lavg, meters) -0.008 0.036 Average number of lanes per approach 0.492 0.000 Cross street is through street (=1 if yes, =0 otherwise) 0.820 0.000 Major T junction (=1 if yes, =0 otherwise) 0.748 0.008 Constant -1.197 0.002 N = 132, LR 2 (prob.) = 135.76 (0.000), Log likelihood = -141.580, chibar2 (prob.) = 341.99 (0.000) Table 1: Severe crash frequency model for Guadalajara (Poisson) Table 2: All crash frequency model for Guadalajara (negative binomial) 18. Factors influencing crash frequencies Counterflow Counterflow lanes were strongly correlated with higher crash frequencies across all our models 19. Factors influencing crash frequencies Street width and intersection size and complexity Metrobus Line 1, Mexico City Road width and complexity of intersections were the most important predictors of crash frequencies. 20. Factors influencing crash frequencies Location of bus lanes Central median Shorter pedestrian crossings Fewer mixed traffic lanes Some 4-way intersections turned into T junctions 21. The safety impact of large blocks For each additional 10 m (30) between signalized intersections: 2% decrease in all crashes 3% increase in severe crashes 22. Wider streets are more problematic for safety Wide, complex intersections can easily become black spots The length of crosswalks is key for pedestrian safety (for each additional meter a pedestrian needs to cross without a median, there is a 2 to 3% increase in pedestrian crashes) Counterflow is the most dangerous bus lane configuration Overview of findings from data analysis 23. A detailed inspection of a street with the objective of identifying safety issues Involves walking along the entire length of a study site (in our case, a BRT corridor) documenting problems related to infrastructure and road user behavior Road safety inspections 24. Fatalities on BRT corridors by road user type Fatalities by Road User Type 0% Pedestrians 54%Car occupants 23% Motorcyclists 10% Bicyclists 5% Other 8% The safest place to be on a bus corridor is inside the bus The most dangerous: pedestrian crossing the avenue 25. Crossing in mid-block 26. Mid-block signalized crosswalk - TransOeste 27. Pedestrian signal timing 28. Pedestrian signal timing 29. Pedestrian signal timing 30. Pedestrian signal timing 31. Pedestrian signal timing 32. Pedestrian signal timing 33. Pedestrian signal timing 34. Pedestrian signal timing 35. Pedestrian signal timing 36. Pedestrian signal timing 37. Pedestrian signal timing 38. Pedestrian signal timing 39. Pedestrian signal timing 40. Pedestrian signal timing 41. BRT Safety recommendations 41 41 SPEED REDUCTIONS: PEDESTRIAN CROSSINGS: INCREASED SIGNAL TIME: To increase safety on Av. das Americas, EMBARQ proposed 42. Speed at Avenida das Amricas 42 43. Speed It is very difficult to control speed through signage and enforcement Streets should be designed for their desired speed 44. Pedestrian bridges Pedestrian bridges rarely work, as pedestrians prefer to cross under them In some cases, this is due to the fear of crime (e.g. research in Cape Town showing pedestrians are concerned of being assaulted or robbed on overpasses) Our research shows they are not effective for safety on typical urban arterials - no significant impact on pedestrian crash frequencies The exception is on high-speed, high-volume roads (e.g. Autopista Norte, Bogota, where locations with pedestrian bridges had four times fewer pedestrian crashes 45. Improving the BRT routes Based on EMBARQs recommendations Stations barriers were improved to avoid unregulated and unsafe pedestrian crossings Install lateral barriers in the platform doors The city is studying to reduce speed limits from 80km/h to 60 km/h in Avenida das Amricas 46. 2011: Signalized mid-block crossing on Eje 2 Oriente, with no traffic calming. Vehicles did not stop for pedestrians. 2012: Speed hump installed before the pedestrian crossing, slowing traffic down and allowing pedestrian to cross safely. More closely spaced signalized intersections can help avoid unsafe speeds But drivers may disregard signals where the only conflict is with pedestrians Speed reductions 47. There is often a considerable difference between how streets are meant to be used and how people actually use them: Jaywalking, including crossing under pedestrian bridges, jumping over guardrails, destroying guardrails Crossing on red (pedestrians) or driving on red (drivers) The key to designing safer streets is understanding road user behavior and how the design of the infrastructure can be better adapted to that Overview of findings from inspections 48. The impact of BRT systems on traffic safety Factors that influence safety on BRT corridors Relationship between safety and operational performance Summary 49. Most safety recommendations involved speed reductions, traffic calming, or additional traffic signals This raises the question of how safety countermeasures would impact the operational performance of a BRT We illustrate the relationship between safety and operational performance with