SOME INSIGHTS ON ROADWAY INFRASTRUCTURE DESIGN FOR SAFE ELDERLY PEDESTRIAN TRAVEL

• Main Author: Venkataraman N. SHANKAR
• Format: Article
• Language: English
• Published: Elsevier 2006-01-01
• Series: IATSS Research
• Subjects: Pedestrians; Bayesian; Frequency; Severity; Logit
• Online Access: http://www.sciencedirect.com/science/article/pii/S0386111214601527

This paper presents insights into the relationship between road infrastructure and elderly pedestrian involvement in traffic accidents. We combine insights from empirical studies involving the probability of a pedestrian accident with insights from studies involving the probability of injuries to elderly pedestrians who are involved in vehicle-pedestrian accidents. The combined insights provide some direction to the methodology for identifying non-motorized improvements for supporting safe elderly travel. The results of the study indicate that after controlling for vehicle volumes, road infrastructure variables posing the greatest risk of pedestrian accidents in urban corridors include the presence of center turning lanes, traffic signal spacing exceeding 0.5 miles and roadway illumination. Center turning lanes indicate the presence of long corridors which may induce elderly pedestrians especially to attempt to cross roadways mid-block using center turning lane sections as refuges. Presence of traffic signals provides reduced pedestrian accident risk if the spacing is less than 0.5 miles. Especially for elderly pedestrians, the availability of protected crossings at signalized intersections is important considering the fact they cannot travel long block lengths in order to use signalized crossings. Presence of continuous roadway lighting decreases elderly pedestrian accident risk. The results also show the greatest impacts on injury severity probabilities are from the occurrence of elderly pedestrian accidents in non-intersection locations. Specifically, if elderly pedestrians are involved in marked crosswalk accidents, the probability of lower severity injury is higher; in contrast, if they are involved in unmarked, non-intersection locations such as mid-blocks, the probability of high severity injury is higher. We obtained these results through the use of Bayesian analysis. Bayesian analysis allows us to use subjective prior information on the distribution of parameters in combination with information from the observed data. The advantage of Bayesian analysis in the assessment of key road variables on safe elderly travel is that we can examine the robustness of results.