A Review of Data Analytic Applications in Road Traffic Safety. Part 1: Descriptive and Predictive Modeling

Area of Science:

• Traffic Safety
• Data Science
• Operations Research

Background:

• Motor vehicle crash risk analysis is fragmented into predictive modeling and route optimization.
• Limited translation exists between these two research streams, hindering progress.
• Current models rarely incorporate near real-time crash risk data.

Purpose of the Study:

• To reduce the burden of data collection and descriptive analytics for crash risk modeling and route optimization.
• To bridge the gap between predictive crash risk models and risk minimization optimization techniques.
• To provide accessible data sources and analytical methods for safer routing.

Main Methods:

• Data-driven bibliometric analysis to identify research streams.
• Review of statistical and machine learning models for crash risk.
• Presentation of publicly available data sources and descriptive analytic techniques.
• Provision of code for data collection and exploration.

Main Results:

• Literature is divided into predictive/explanatory models and optimization techniques.
• Limited transfer of findings between the two identified research streams.
• Near real-time crash risk is infrequently considered in current models.

Conclusions:

• Accessible data and descriptive analytics can facilitate safer routing.
• Integrating predictive insights into optimization models requires addressing the real-time risk data gap.
• Further research should focus on bridging the identified research streams for enhanced traffic safety.