Post encroachment time-based behavioral analysis of vehicle-pedestrian interactions at unsignalized midblock crosswalks

• Context‐specific PET thresholds derived for urban, arterial, and mixed traffic using survival analysis. • Key PET factors: vehicle type, yield, own path, crossing speed, conflict type, curb time, traffic speed, perceived volume. • Heavy vehicles posed highest risk; SUVs showed greater severity than two-wheelers. • Study outcomes support PET-based safety evaluation for mixed traffic in developing contexts. Most existing research on vehicle–pedestrian interactions often overlooks the key factors influencing Post-Encroachment Time (PET), a critical surrogate safety measure. Risk classification thresholds are typically based on assumptions or percentile-based approaches, while severity assessments rely on either conflict or crash data in isolation, limiting the comprehensiveness of analysis. This study addresses these gaps by exploring a range of behavioral, environmental, and interaction-related variables to identify determinants of PET. A data-driven method was proposed to derive context-sensitive PET thresholds through survival analysis, and an integrated severity analysis framework was developed by combining conflict data with AIS-based crash modifiers for a realistic estimate of injury risk. Multiple Linear Regression identified Curb Time, Vehicle Type, Yielding Behavior, Own Path, Conflict Type, Crossing Speed, Traffic Speed, and Perceived Volume as key PET predictors. Survival analysis yielded PET thresholds of 1.21 s and 2.45 s for urban road, and 1.71 s and 2.98 s for arterial road, enabling categorization into High, Medium, and Low-Risk interactions. Generalized thresholds of 1.34 s and 2.88 s are recommended when site-specific data is unavailable. Severity analysis revealed the highest risks for Bus-Trucks (16.90 % Extreme Severity), followed by SUV-Cars (59.22 % High to Very High), and Two-Wheelers (56 % Low to Moderate). The findings recommend adopting context-specific PET thresholds for more accurate risk classification and enforcing stricter yielding regulations and speed limits near midblock crosswalks—particularly in areas with a high presence of heavy vehicles. The methodology is adaptable to other urban contexts, offering practical implications for proactive safety planning, especially in developing countries.