Some of the fundamental problems with design information stem from the nature of the primary source of data on traffic crashes. Accident reported filed by police agencies lack the detail and precision necessary to correlate accidents to specific points on the roadway. Consequently, it is not possible to develop predictive models for accidents or even assess the potential benefit of improvements. Further, resources for improving safety are limited, particularly in lieu of the completion of the interstate system and a massive need to address major maintenance aspects of the nation's highway infrastructure. There has also been significant changes in the nature of the vehicle fleet operating on the nations highway and the attitudes of drivers that affect the safety of the highway. Further new concepts have evolved for tailoring the highway design to its intended performance requirements. In the absence of better and even in some cases basic information on the safety impacts of design decisions and the limits on available resources, traditional practices are continued.
Abstract