Naturalistic driving study data applied to road infrastructure: A systematic review.

Area of Science:

• Road safety research
• Transportation engineering
• Human factors in driving

Background:

• Naturalistic driving studies (NDS) have significant potential for characterizing road infrastructure's influence on driving behavior.
• A systematic review evaluated best practices in applying NDS data to road infrastructure research.

Purpose of the Study:

• To systematically review and evaluate the objectives, data processing, and analysis methods in NDS research focusing on road infrastructure.
• To assess the quality and methodologies of existing NDS studies related to road infrastructure.

Main Methods:

• Systematic literature search following PRISMA guidelines across seven databases.
• Analysis of 53 peer-reviewed studies based on their primary infrastructure research aims.
• Quality assessment using amended NHLBI templates; data source and processing methods were cataloged.

Main Results:

• Studies focused on curves, intersections, ramps, and work zones, with 74% assessed as 'Good' quality.
• External video was the primary data source (38%), often supplemented by satellite and government data.
• 91% of studies required significant data preparation, including manual video coding, to link datasets and extract variables.

Conclusions:

• NDS methods are adaptable but require significant effort for non-NDS data and video coding.
• Reducing computational costs in sample processing can enhance NDS research on road infrastructure.
• Integrating road infrastructure factors into NDS within a Safe System approach is recommended for future research planning.