Empirical Bayes Procedure for Ranking Sites for Safety Investigation by Potential for Safety Improvement

Abstract

The identification of sites requiring investigation for possible safety treatment is one of the most important aspects of infrastructure safety management and has been the subject of considerable research aimed at improving the efficiency of the process. The more recent techniques use the empirical Bayes (EB) method for estimating the safety of specific sites. A refinement of the EB method that is conceptually sound and inherently simple to apply is the focus of this research. With this refinement, the EB estimate of the safety of a site is compared with its expected safety to determine the potential for safety improvement, which forms the basis for ranking sites for safety investigation. A comparative evaluation is provided of the proposed method against other EB methods and the more conventional ones, with data for signalized intersections and two-lane rural highway sections. The refined EB method is shown to be comparatively efficient.

Get full access to this article

View all access options for this article.

References

Hauer

, and Persaud

. Problem of Identifying Hazardous Locations Using Accident Data. In Transportation Research Record 975, TRB, National Research Council, Washington, D.C., 1984, pp. 26–43.

Hauer

Empirical Bayes Approach to the Estimation of “Unsafety”: The Multivariate Regression Method. Accident Analysis and Prevention, Vol. 24, No. 5, 1992, pp. 457–477.

Higle

J. L.

, and Witkowski

J. M.

. Bayesian Identification of Hazardous Locations. In Transportation Research Record 1185, TRB, National Research Council, Washington, D.C., 1988, pp. 24–36.

Hauer

Identification of “Sites with Promise.” In Transportation Research Record 1542, TRB, National Research Council, Washington, D.C., 1996, pp. 54–60.

Persaud

B. N.

, Cook

, and Kazakov

. Demonstration of New Approaches for Identifying Hazardous Locations and Prioritizing Safety Treatment In Proc., 7th International Conference: Traffic Safety on Two Continents, Lisbon, Portugal, September 22–24, 1997.

McGuigan

D. R. D.

The Use of Relationships Between Road Accidents and Traffic Flow in ‘Black-Spot’ Identification. Traffic Engineering and Control, Aug.-Sept. 1981, pp. 448–453.

Jorgensen

N. O.

Statistical Detection of Accident Blackspots. OTAPIARC 11th International Study Week in Transportation and Safety. Brussels, 1972.

Tarko

, Sinha

, and Farooq

. Methodology for Identifying Highway Safety Problem Areas. In Transportation Research Record 1542, TRB, National Research Council, Washington, D.C., 1996, pp. 49–53.

Persaud

B. N.

Estimating the Accident Potential of Ontario Road Sections. In Transportation Research Record 1327, TRB, National Research Council, Washington, D.C., 1991, pp. 47–54.

10.

Baker

R. J.

, and Nelder

J. A.

. The GLIM System-Release 3. Rothamsted Experimental Station, Harpenden, U.K., 1978.