Unraveling Pedestrian Fatality Patterns: A Comparative Study with Explainable AI

2025-03-22

AI Summary

This study identifies key causal disparities underlying the highest and lowest pedestrian fatality rates across U.S. states. Method: Leveraging the Fatality Analysis Reporting System (FARS) database, we develop a novel cross-state comparative explainable artificial intelligence (XAI) framework: SMOTE addresses severe class imbalance; an ensemble of models—including XGBoost—is trained and interpreted via SHAP to quantify feature contributions at a policy-relevant level. Contribution/Results: XGBoost achieves 98% balanced accuracy. Critical high-risk factors include mid-block roadway locations, low visibility conditions, older pedestrian age, and alcohol/drug impairment. Findings reveal substantial regional heterogeneity in pedestrian fatality drivers, informing differentiated interventions—such as targeted lighting improvements, infrastructure upgrades, and precision enforcement—and providing data-driven, evidence-based decision support for traffic fatality prevention.

Technology Category

• Humans and AI: Explainable AI (XAI) for Human Understanding
• Reasoning under Uncertainty: Causality
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Road fatalities pose significant public safety and health challenges worldwide, with pedestrians being particularly vulnerable in vehicle-pedestrian crashes due to disparities in physical and performance characteristics. This study employs explainable artificial intelligence (XAI) to identify key factors contributing to pedestrian fatalities across the five U.S. states with the highest crash rates (2018-2022). It compares them to the five states with the lowest fatality rates. Using data from the Fatality Analysis Reporting System (FARS), the study applies machine learning techniques-including Decision Trees, Gradient Boosting Trees, Random Forests, and XGBoost-to predict contributing factors to pedestrian fatalities. To address data imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) is utilized, while SHapley Additive Explanations (SHAP) values enhance model interpretability. The results indicate that age, alcohol and drug use, location, and environmental conditions are significant predictors of pedestrian fatalities. The XGBoost model outperformed others, achieving a balanced accuracy of 98 %, accuracy of 90 %, precision of 92 %, recall of 90 %, and an F1 score of 91 %. Findings reveal that pedestrian fatalities are more common in mid-block locations and areas with poor visibility, with older adults and substance-impaired individuals at higher risk. These insights can inform policymakers and urban planners in implementing targeted safety measures, such as improved lighting, enhanced pedestrian infrastructure, and stricter traffic law enforcement, to reduce fatalities and improve public safety.

Problem

• Identifying key factors in pedestrian fatalities using XAI
• Comparing high and low fatality states with machine learning
• Analyzing age, substance use, and environment as risk predictors

Innovation

• Explainable AI for pedestrian fatality analysis
• Machine learning with SMOTE for data balance
• SHAP values for model interpretability enhancement