Capturing Road-Level Heterogeneity in Crash Severity on Two-Lane Rural Highways: A Multilevel Mixed-Effects Approach

2025-08-13

AI Summary

To address the neglect of unobserved heterogeneity across road segments in modeling crash severity on rural two-lane highways, this study proposes a random-coefficients multilevel mixed-effects model. The model integrates multivariate predictors—including driver characteristics, roadway environment, and traffic flow—and allows key covariate effects to vary randomly across segments to capture latent segment-level heterogeneity. Compared with single-level generalized linear models and random-intercept models, the proposed approach significantly improves predictive performance: classification accuracy increases from 0.62 to 0.71, recall rises from 0.32 to 0.63, and the area under the ROC curve (AUC) reaches 0.775. These results demonstrate the superiority and practical utility of multilevel modeling in uncovering localized risk mechanisms and supporting context-sensitive safety interventions.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Machine Learning: Multi-instance/Multi-view Learning
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Security and Privacy: Large-scale security measurements
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Accurately modeling crash severity on rural two-lane roads is essential for effective safety management, yet standard single level approaches often overlook unobserved heterogeneity across road segments. In this study, we analyze 19 956 crash records from 99 rural roads in Iran during recent four years incorporating crash level predictors such as driver age, education, gender, lighting and pavement conditions, along with road level covariates like annual average daily traffic, heavy-vehicle share and terrain slope. We compare three binary logistic frameworks: a single level generalized linear model, a multilevel model with a random intercept capturing latent road level effects (intraclass correlation = 21 %), and a multilevel model with random coefficients that allows key predictor effects to vary by road. The random coefficient model achieves the best fit in terms of deviance, AIC and BIC, and substantially improves predictive performance: classification accuracy rises from 0.62 to 0.71, recall from 0.32 to 0.63, and AUC from 0.570 to 0.775. Results from 200 simulation runs reveal notable variability in slopes for pavement and lighting variables, underscoring how local context influences crash risk. Overall, our findings demonstrate that flexible multilevel modeling not only enhances prediction accuracy but also yields context-specific insights to guide targeted safety interventions on rural road networks.

Problem

• Model crash severity on rural two-lane roads accurately
• Address unobserved heterogeneity across road segments
• Improve predictive performance for safety interventions

Innovation

• Multilevel mixed-effects modeling for crash severity
• Random coefficients to capture road-level variability
• Improved prediction accuracy with context-specific insights