A hierarchical bayesian peak over threshold approach for conflict-based before-after safety evaluation of leading pedestrian intervals. (November 2020)

• Record Type: Journal Article
• Authors: Guo, Yanyong; Sayed, Tarek; Zheng, Lai
• Is Part Of: Accident analysis and prevention. Volume 147(2020)
• Publication Date: 2020-11
• Subjects: Traffic conflicts -- Before-after study -- Hierarchical bayesian model -- Extreme value theory -- Leading pedestrian intervals; Accidents -- Prevention -- Periodicals; Accident Prevention -- Periodicals; Accidents -- Prévention -- Périodiques; 363.106
• DOI: 10.1016/j.aap.2020.105772
• Languages: English
• ISSNs: 0001-4575

Abstract

Highlights: A conflict-based EVT before-after safety study is used to evaluate treatment effect of LPI. A hierarchical Bayesian peak over threshold approach is proposed to model conflict extremes. A hierarchical Bayesian structure is used to incorporate covariates and unobserved heterogeneity. Traffic conflicts of different sites and periods are combined to evaluate the treatment effect. Significant safety improvements are found after the implementation of LPI. Abstract: A hierarchical Bayesian peak over threshold (POT) approach is proposed for conflict-based before-after safety evaluation of Leading Pedestrian Intervals (LPI). The approach combines traffic conflicts of different sites and periods to develop a uniform generalized Pareto distribution (GPD) model for the treatment effect estimation. The hierarchical structure has three levels, a data level that consists of modeling the traffic conflict extremes through the POT approach, a latent process level that relates GPD parameters of the data level to certain covariates, and a prior level with prior distributions to characterize the latent process. The approach was applied to a before-after (BA) safety evaluation of leading pedestrian interval (LPI) in Vancouver, BC. Pedestrian-vehicle traffic conflicts were collected from treatment and control sites during the before and after periods using an automated computer vision analysis technique. The treatment effect was measured by the best fitted GPD model with the calculation of the odds ratio (OR). The overall treatment effect varies from 18.1%–20.9% in terms of reduction in the estimated extreme-serious conflicts. The treatment effect indicates a considerable improvement in pedestrian safety after the implementation of the LPI, and the consistent results demonstrate a reliable BA safety evaluation. As such, the proposed approach is recommended as a promising tool for BA safety studies, particularly in cases where collision data is limited.