Integrated traffic conflict model for estimating crash modification factors

Area of Science:

• Transportation Engineering
• Traffic Safety Analysis
• Road Safety Treatments

Background:

• Crash modification factors (CMFs) are crucial for road safety but traditionally rely on observational crash data, which can be limited and lack causal insights.
• Existing methods like empirical Bayes (EB) before-and-after analyses improve precision but require extensive data and struggle with certain treatments.
• Observational CMFs do not provide a behavioral rationale for safety improvements, necessitating alternative approaches.

Purpose of the Study:

• To develop and investigate an integrated approach for estimating CMFs using simulated traffic conflicts formally linked to observed crashes.
• To address the limitations of traditional observational methods in CMF estimation, particularly regarding causality and data requirements.
• To compare the effectiveness of the proposed integrated method with established empirical Bayes (EB) crash-based before-and-after analysis.

Main Methods:

• Utilized simulated traffic conflicts, a surrogate safety measure, and formally linked them to observed crash data.
• Developed an integrated method to estimate CMFs by combining simulated conflict data with crash records.
• Compared the CMF estimates from the integrated method against those derived from empirical Bayes (EB) before-and-after analysis for a specific road safety treatment.

Main Results:

• The proposed integrated method for estimating CMFs yielded results that closely aligned with estimates from the traditional empirical Bayes (EB) crash-based before-and-after analysis.
• The study demonstrated the feasibility of linking simulated traffic conflicts with observed crashes to derive reliable CMFs.
• The integrated approach shows promise in providing a more causal perspective on safety treatments.

Conclusions:

• The integrated method of using simulated conflicts linked to observed crashes offers a viable and promising alternative for estimating CMFs.
• This approach can potentially overcome data limitations and provide a stronger causal basis for road safety treatment effectiveness.
• The findings support the use of integrated methods for more robust and insightful road safety analysis.