A multivariate analysis of crash and naturalistic driving data in relation to highway factors
Authors: T. J. Gordon, L. P. Kostyniuk, P. E. Green, M. A. Barnes, D. F. Blower, S. E. Bogard, A. D. Blankespoor, D. J. LeBlanc, B. R. Cannon, S. B. McLaughlin
This report documents the second phase of a two-phase project under the Transportation Research Board’s second Strategic Highway Research Program (SHRP 2) Safety Project S01C. A primary part of this work involved conducting a multivariate analysis of crash and naturalistic driving data in relation to highway factors. A geographic information system (GIS) framework was used as the basis for fusing multiple information sources to analyze road departure crash risk. A major goal was to use this method to support formulation and validation of crash surrogates. Two analytical models developed in the study focus on the statistical relationship between surrogate measures of crashes and actual crashes and on the formulation of exposure-based risk measures using surrogate measures. The report also describes three exploratory studies that illustrate the value of the geospatial approach taken. The results of each exploratory study suggest that the combination of naturalistic, crash, and highway data provides a rich data resource for many types of research. This report provides valuable background information to highway safety analysts seeking to use the data that will be made available from the SHRP 2 naturalistic driving study.