Exploration of Some Key Issues in Developing and Applying Crash–Conflict Models for Signalized Intersections

Rajeswaran, Thanushan; Persaud, Bhagwant; Anarkooli, Alireza Jafari · 2023 · Crossref

DOI: 10.1177/03611981221148483

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the challenge of evaluating road safety strategies under Vision Zero programs, where historical crash data may be insufficient. It focuses on developing and applying crash-conflict models that utilize surrogate safety measures, specifically traffic conflicts derived from microsimulation, to predict crash frequencies at signalized intersections. The research investigates key issues including the inclusion of vehicle speed as a predictor, the definition of conflict severity, model transferability across jurisdictions, and the applicability of these models for estimating Crash Modification Factors (CMFs). The methodology involved microsimulating 91 four-legged signalized intersections in Toronto using SYNCHRO and PTV VISSIM software to generate vehicle trajectories. These trajectories were analyzed using the Surrogate Safety Assessment Model (SSAM) to identify vehicle-vehicle conflicts based on Time to Collision (TTC) and Post Encroachment Time (PET) thresholds. Crash prediction models were developed using a Generalized Linear Modeling approach with a Negative Binomial error structure, incorporating conflict counts, average and maximum speeds of conflicting vehicles, and a calculated risk score. To assess transferability, the models were calibrated to data from 13 intersections in York Region. Additionally, the study tested the models' ability to estimate CMFs by simulating changes in left-turn phasing at 10 Toronto intersections. The results indicate that incorporating speed variables significantly improves model performance compared to using conflict counts alone. Models utilizing the 2.5-second PET threshold and average speed yielded the best fit, with speed coefficients showing that crash frequency increases with higher speeds. The risk score approach confirmed that injury crashes are associated with more severe conflicts than total crashes, though its overdispersion parameters were slightly higher than other models. Regarding transferability, models without speed variables performed best when calibrated to the York Region dataset, though all models met acceptable goodness-of-fit criteria. The study also demonstrated that the improved crash-conflict models could effectively estimate CMFs for signal phasing changes, producing results consistent with empirical Bayes before-after studies. The significance of this work lies in validating the use of microsimulation-derived conflicts and speed metrics for safety assessment. The findings confirm that including speed enhances the predictive power of crash-conflict models and supports their use for estimating site-specific CMFs. This approach offers a viable alternative for evaluating safety treatments where crash data is sparse, although the authors note that further research is needed to optimize model formulations and simulation parameters for broader application.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• induced exposure
• regulatory evaluation
• incidence prevalence

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: crash risk outcomes