SWOV homepage
Nederlands  

Searching for quantitative relations between road design and road safety

 

In various countries it is standard practice to quantify the relationship between geometric, engineering elements in road design and the chance of accidents occurring. Given this relationship, road designers are trying to optimize their design. In the United States, this approach has been summarized in the Interactive Highway Safety Design Model (IHSDM) under the motto 'designing highways with safety in mind'. Within Europe, this approach has been greatly developed in Germany, Sweden, Portugal, and the United Kingdom.

This approach to road designing would seem to be relevant in developing a sustainably safe traffic system. To see if this is so, an exploratory literature study of the relevant, foreign research has been carried out (R-98-49).

 

The research questions were:

 

Structure

The study is structured according to the usual arrangement for guidelines in road design. These are: alignment (tangents, curves, and relation design), cross-section, and intersections (including roundabouts). Special attention has been paid to the size of the sample, (external) factors that could disturb the results, and the robustness of any statistical relationships found. A lot of use has consciously been made of those studies that have been peer-reviewed, to ensure their quality.

 

A number of accident models developed in the United States appear to be about to be used in IHSDM. Here, the experience gained in the 1980s is being applied. This knowledge concerns mainly rural, single-carriageway roads. This knowledge was already available to American designers in generally accessible publications. This knowledge, however, was not yet integrated in the existing guidelines.

Models for junctions are available in some European countries. Their application in design guidelines, or road design, has not yet advanced very far.

 

The geometric design of sustainably safe roads would benefit from being able to quantify the expected safety effects of various alternative types of lay out. Safety and accident models assist for this, as do overviews of reported effects. Existing foreign models should be adapted to the circumstances in the Netherlands. Practical application can only be successful if the model details fit the design variations at hand.

SWOV Research Activities 11 - March 1999

Knowledge base