Study Details

Study Title: Linear and Nonlinear Safety Intervention Models: Novel Methods Applied to the Evaluation of Shoulder Rumble Strips

Authors: El-Basyouny and Sayed

Publication Date:JAN, 2012

Abstract: The most common practice in safety evaluation utilizes a longitudinal (before-after) study design with comparison groups to estimate the effects of a safety countermeasure. These longitudinal evaluations base their results on actual changes that have occurred over a period of time extending from the before condition to the after condition. Recent safety evaluation research has advocated the use of linear intervention models developed within a Hierarchical full Bayes context. These models acknowledge that the safety treatment (intervention) effects do not occur instantaneously but are spread over future time periods. Despite the demonstrated advantages of such models, there is still a lack of complete understanding of how the implemented countermeasures affect safety at the treated locations in terms of novelty, direct and indirect effects. This paper proposes a novel nonlinear intervention model in an attempt to provide a better understanding of how safety countermeasures work. The model is known as the 'Koyck' intervention model. To demonstrate its capabilities, the linear and nonlinear (Koyck) models are applied to estimate the effectiveness of the installation of shoulder rumble strips in a number of highway segments in the province of British Columbia, Canada. For both models, the treatment effectiveness was measured as a function of post treatment time providing analysts a better representation of the treatment impact in subsequent years after installation. Moreover, the treatment impact was decomposed into direct and indirect effects. The direct treatment impact enables the analysts to assess the effectiveness of the countermeasures apart from local (site-related) factors. In addition to providing the best fit, the nonlinear 'Koyck' model has provided valuable insight into the effectiveness of shoulder rumble strips showing an immediate 24.9% reduction after one year which decreased with time resulting in a permanent treatment impact of 19.2%. The results of this study can have a significant impact on the economic evaluation of safety programs and countermeasures. Given the way future benefits are discounted to present values, the results of using a collision modification factor that changes with time can significantly affect the estimation of the cost effectiveness of safety countermeasures.

Study Citation: El-Basyouny, K. and Sayed, T. "Linear and Nonlinear Safety Intervention Models: Novel Methods Applied to the Evaluation of Shoulder Rumble Strips." Transportation Research Record: Journal of the Transportation Research Board, No. 2280, Transportation Research Board of the National Academies, Washington, D.C., 2012, pp. 28-37. DOI: 10.3141/2280-04

Study Report: Download the Study Report Document


CMFs Associated With This Study

Category: Shoulder treatments

Countermeasure: Install shoulder rumble strips

CMF CRF(%)QualityCrash TypeCrash SeverityRoadway TypeArea Type
0.82217.84 StarsRun off roadAllNot SpecifiedRural