Bayesian Ranking of Sites for Engineering Safety Improvements: Decision Parameter, Treatability Concept, Statistical Criterion, and Spatial Dependence

The objective of this study was to explore some of the issues raised in recent roadway safety studies regarding ranking methodologies in light of the recent statistical development in space-time generalized linear mixed models (GLMM). These issues include: identifying appropriate roles and objectives of ranking methodologies in achieving the overall strategy of improving roadway safety; understanding the context and process in which the methodology is to be applied; developing implementation strategies with various levels of data requirements; identifying target users and considering their resource and data constraints; choosing ranking approaches that can properly handle the highly stochastic nature of the crash event due to low and uneven exposure by site; selecting decision parameters for ranking, e.g., crash frequency rate or crash cost rate in number of crashes or US$ 1000, per vehicle miles traveled or per vehicle entry; making the use of treatability concept to assess known traffic, geometric, and other explanatory covariate effects and determining the need to include spatial, temporal, and unobserved site heterogeneities in such an assessment;

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01002232
  • Record Type: Publication
  • Files: TRIS, ATRI
  • Created Date: Jul 15 2005 2:33PM