This report provides a comprehensive overview of current understanding of the relationships between safety and design. It was initiated on the premise that the safety of a highway is intrinsically based in its design. It is clear that design decisions related to the horizontal and vertical alignment of a highway influence the way vehicles operate on a roadway and consequently the risks associated with the use of the roadway. Less apparent - but equally important to safety - are decisions related to the selection of cross-sectional features and the provisions for the use of ancillary hardware in the roadway environment. The realities of limited resources, environmental concerns, land use constraints, and other factors make it difficult to design roadways which maximise safety. The original project objective was to develop guidelines for selecting highway design elements and parameters which would maximise safety benefits. It was believed that sufficient research had been conducted to understand the effects of geometric and traffic features on safety, but that the results had not been synthesised into a unified document and correlated with current design practices. The research was intended to address geometric, cross-sectional, and roadside design elements for all roadway types, environments, and traffic situations. The research began with an in-depth literature review and state surveys. The literature revealed a variety of models and data relating design and safety. These were translated to a similar format and categorised by feature addressed. Contacts with highway designers obtained information on the procedures used and types of design issues faced. The contractor compiled the findings of these efforts in an interim report which indicated that while sound research efforts had led to a greatly improved understanding of the effects of design on safety, some relationships were not well defined and significant gaps remained in the knowledge about these effects. For example, many of the studies, that are used as the basis for current practices, included only a limited set of conditions, reflected the biases of particular state practices, and involved small samples. A significant deficiency exists in the understanding of the combined and interactive effects of multiple design features (e.g., the relative safety of curves on a downgrade). Further, the contacts with states indicated that there was a wide variation in practices and only limited definition of threshold values for limiting design conditions. These conclusions led to the decision to refocus the objectives of the project to developing recommendations for a research program to fill the gaps in the current state of the practice. The report outlines eight research plans formulated to address the existing gaps in the state of the practice. (A)
Abstract