Citation

Sun X, Li Y, Magri D, Shirazi HH. Transp. Res. Rec. 2006; 1950: 55-64.

Copyright

(Copyright © 2006, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Predicting highway safety performance in quantitative terms on the basis of a set of highway geometric design and traffic control features is the core of the ongoing "Highway Safety Manual" (HSM) development. While great efforts have been made to develop safety performance models for different types of highways, evaluating the applicability of the models with crash data from states that were not used in the original model development is equally important. This step is critical in future HSM applications because of many noticeable differences in crash data recording practice and driving behaviors across the various states. Potential application concerns need to be evaluated fully and addressed before the manual is officially published. This paper presents the application experience of the safety prediction models and methodology for segments of two-lane rural roads as proposed for inclusion in Part III of the HSM. The study used data for state routes in Louisiana. The results indicate that the prediction models performed reasonably well: there were small differences between the observed and predicted crash frequencies, with calibration parameters applied, for all segments on rural two-lane highways in Louisiana. The results at different analysis levels (from individual segments to a network of segments) are summarized, and several application issues such as the level of effort versus degree of accuracy, simplification of the calibration process, and the need for a reliability measurement are discussed.