Citation

Gargoum SA, El-Basyouny K. Accid. Anal. Prev. 2019; 136: e105395.

Affiliation

Department of Civil and Environmental Engineering, University of Alberta, Edmonton, AB, T6G 1H9, Canada. Electronic address: basyouny@ualberta.ca.

Copyright

(Copyright © 2019, Elsevier Publishing)

DOI

10.1016/j.aap.2019.105395

PMID

31877448

Abstract

Stopping Sight Distance (SSD) is the distance defined in most highway design guides as the distance required by drivers to safely come to a complete stop in case of an emergency. Accordingly, design guides define theoretical values for SSD and recommend that these requirements are satisfied at all points along a highway corridor. SSD is estimated as a function of speed, driver reaction time, and deceleration rate, which are all factors that vary by both driver and driving conditions. Despite the anticipated uncertainty in those variables, they are all modelled deterministically. Unfortunately, this is an inaccurate assumption and provides no information about the extent to which roads designed to meet SSD requirements are able to satisfy road user demand for SSD. Design guides also fail to provide information about the impact a segment that fails to meet driver needs has on safety. To overcome those limitations, this paper assesses the ability of existing roads to satisfy stochastically modelled road user demand for SSD. The Available Sight Distance (ASD) was first quantified for a group of top crash-prone segments, and a Monte Carlo Simulation was used to model demand for SSD. The proportion of the test highways that failed to meet driver demands for SSD was then quantified by comparing the ASD to the required SSD at different levels of driver demand. Furthermore, the paper also compares the safety performance between regions that meet SSD and those that fail to do so. Among other findings, the paper shows that, on average, 6.8 % of the length of the test segments are noncompliant to the SSD demands of 70 % of the driving population. On the other hand, the average percent noncompliance for 30 % of the driving population (the 30 % with limited abilities) was 12.1 %. It was also found that, on average, crash rates in the noncompliant regions were two to three times higher than those in the compliant regions at the 70 % level (i.e., in regions that fail to meet the SSD demands of 70 % of the driving population).

Copyright © 2019 Elsevier Ltd. All rights reserved.

Language: en

Keywords

Available sight distance; Crash rates; LiDAR; Road user ability; Stochastic assessment; Stopping sight distance demand; Stopping sight distance limitations