|

Safety Evaluation of the Variable Speed Limit System (VSLS) in British Columbia

Author(s): Elesawey, Ludwar, Sengupta, Babineau

Slidedeck Presentation Only (no paper submitted):

7A - Elesawey

Abstract:

Posted speed limits are established for ideal road conditions. If roadway conditions are less than ideal, such as during heavy rain, snow or black ice, regular speed limit signs may not display an appropriate speed limit for those conditions. Drivers are expected to adjust their speed to the conditions. However, even when adverse weather conditions are recognized by drivers, they may not know how much to reduce their speed. The VSLS uses real-time collected weather and traffic data to inform drivers of the maximum acceptable speed based on current conditions. To undertake a safety evaluation of the VSLS installed on the following three corridors in BC in June 2016:
* Highway 1 from Perry River to Revelstoke
* Highway 5 from Portia Interchange to the former Toll Plaza
* Highway 99 from Squamish to Function Junction, Whistler An Empirical Bayes (EB) method was utilized to undertake the safety assessment. The method made use of Safety Performance Functions (SPFs) and a comparison group to account for different confounding factors. The evaluation was undertaken for each corridor separately due to the significant differences in attributes of the three corridors. Early evaluation results showed insignificant increases in winter serious collisions for Highway 1 and Highway 99. This uncertainty indicates inconclusive results. On the other hand, a significant reduction in winter serious collisions was determined for Highway 5. * The safety impact differed significantly from one corridor to another.
* The most pronounced safety benefits were observed on Highway 5 which experienced the most severe winter events.
* The weather-based sub-system of the VSLS on Highway 99 was not operating adequately and in numerous instances the speed limit was selected manually. As such, the assessment of the effect of VSLS on safety may be erroneous and misleading. * The after-implementation period is too short (i.e. only one season) and hence the results are considered preliminary.
* The safety evaluation will be redone after the collision data of the current winter season becomes available.'