Time series modeling in traffic safety research.

Area of Science:

• Transportation Science
• Statistical Modeling
• Time Series Analysis

Background:

• Statistical models are established for traffic safety (crash) data analysis.
• Time series techniques are underrepresented due to data collection challenges and methodological gaps.

Purpose of the Study:

• To provide an overview of the state of the art in time series models for traffic safety research.
• To guide analysts on fundamental techniques and considerations in time series modeling.
• To explore future opportunities and newer modeling techniques, including computational intelligence.

Main Methods:

• Review of existing literature on time series models in traffic safety.
• Discussion of classic time series modeling techniques and considerations.
• Exploration of computational intelligence models for handling large datasets.

Main Results:

• Identifies the growing importance of time series models with high-resolution traffic safety data.
• Highlights the need for guidance on methodology and application.
• Presents computational intelligence as a promising approach for large-scale data analysis.

Conclusions:

• Time series models offer significant potential for advancing traffic safety research.
• Guidance and exploration of advanced techniques are crucial for effective implementation.
• This work provides a framework for understanding safety trends and informing policy.