Detection of Safety Deterioration in a Series of Accident Counts

Abstract

The task is to develop a workable procedure with the object of detecting sites where the mean accident frequency has increased over time to more than what can be attributed to changes in traffic or to general trends. Two kinds of increase are to be detected: a steady but gradual deterioration and a sudden increase in mean accident frequency at an unknown time. The procedure is to be applied to many sites, with the detected sites subsequently examined in detail. Both the gradual and the sudden increase in accident frequency are to be detected by a statistical test of significance for the difference between the means of two Poisson random variables. The software for doing such tests of significance has been developed and is available. An identification procedure is suggested to take account of (a) the number of sites that can be practically subjected to detailed examination and (b) the marginal yield of correctly identified sites.

Get full access to this article

View all access options for this article.

References

Box

G. E. P.

, and Tiao

G. C.

Intervention Analysis with Applications to Economic and Environmental Problems. Journal of American Statistical Association, Vol. 70, 1975, pp. 70–79.

Harvey

A. C.

, and Durbin

The Effects of Seat Belt Legislation on British Road Casualties: A Case Study in Structural Time Series Modelling. Royal Statistical Society, 1986.

Ledolter

, and Chan

K. S.

Evaluating the Impact of the 65 mph Maximum Speed Limit on Iowa Rural Interstates. The American Statistician, Vol. 50, No. 1, Feb. 1996, pp. 79–85.

Hauer

, The Effect of Resurfacing on the Safety of Two-Lane Rural Roads in New York State. In Transportation Research Record 1467, TRB, National Research Council, Washington, D.C., 1994, pp. 30–37.