Comprehensive Analysis of Traffic Safety and Collision Prediction: A Data-Driven Approach to Transportation Risk Assessment

Abstract

This research presents a comprehensive examination of vehicular transportation safety through the lens of predictive analytics and data-driven methodologies. The study investigates collision patterns, traffic flow dynamics, and risk assessment frameworks within modern transportation systems. Through systematic literature review and bibliometric analysis, we identify two primary research streams: (a) descriptive and predictive modeling approaches that utilize historical data to understand collision patterns and risk factors, and (b) prescriptive optimization methods focused on route selection and hazard mitigation strategies. Our analysis encompasses 856 relevant publications, examining various data collection methodologies, analytical techniques, and modeling approaches used in transportation safety research. The findings reveal significant relationships between driver behavior, environmental conditions, infrastructure characteristics, and collision risk. Key contributing factors include driver fatigue, distracted driving behaviors, roadway geometry, weather conditions, and traffic flow patterns. This study provides a framework for integrating diverse data sources and analytical methods to enhance transportation safety prediction and intervention strategies.