Traffic crashes are a leading cause of death and injury worldwide, with far-reaching societal and economic consequences. To effectively address this global health crisis, researchers and practitioners rely on the analysis of crash data to identify risk factors, evaluate countermeasures, and inform road safety policies. This systematic review synthesizes the state of the art in road crash data analysis methodologies, focusing on the application of statistical and machine learning techniques to extract insights from crash databases. We systematically searched for peer-reviewed studies on quantitative crash data analysis methods and synthesized findings by using narrative synthesis due to methodological diversity. Our review included studies spanning traditional statistical approaches, Bayesian methods, and machine learning techniques, as well as emerging AI applications. We review traditional and emerging crash data sources, discuss the evolution of analysis methodologies, and highlight key methodological issues specific to crash data, such as unobserved heterogeneity, endogeneity, and spatial–temporal correlations. Key findings demonstrate the superiority of random-parameter models over fixed-parameter approaches in handling unobserved heterogeneity, the effectiveness of Bayesian hierarchical models for spatial–temporal analysis, and promising results from machine learning approaches for real-time crash prediction. This survey also explores emerging research frontiers, including the use of big data analytics, deep learning, and real-time crash prediction, and their potential to revolutionize road safety management. Limitations include methodological heterogeneity across studies and geographic bias toward high-income countries. By providing a taxonomy of crash data analysis methodologies and discussing their strengths, limitations, and practical implications, this paper serves as a comprehensive reference for researchers and practitioners seeking to leverage crash data to advance road safety.

Accurate geospatial data are essential for intelligent transportation systems and automated reporting applications, as location precision directly impacts safety analysis and decision-making. GPS devices are now routinely employed by law enforcement officers when filing vehicle crash reports, yet our investigation reveals that significant data quality issues persist. The high apparent precision of GPS coordinates belies their actual accuracy as we find that approximately 20% of crash sites need correction—results consistent with existing research. To address this challenge, we present a novel credibility scoring and correction algorithm that leverages a state-of-the-art multimodal large language model (LLM) capable of integrated visual and textual reasoning. Our framework synthesizes information from structured coordinates, crash diagrams, and narrative text, employing advanced artificial intelligence techniques for comprehensive geospatial validation. In addition to the LLM, our system incorporates open geospatial data from Overture Maps, an emerging collaborative mapping initiative, to enhance the spatial accuracy and robustness of the validation process. This solution was developed as part of research leading to a patent for autonomous vehicle routing systems that require high-precision crash location data. Applied to a dataset of 5000 crash reports, our approach systematically identifies records with location discrepancies requiring correction. By uniting the latest developments in multimodal AI and open geospatial data, our solution establishes a foundation for intelligent data validation in electronic reporting systems, with broad implications for automated infrastructure management and autonomous vehicle applications.