In this report three subjects were addressed: (1) an efficiently applicable calculation method; (2) selection of `likely' research variables; and (3) the possibilities for an interactive analysis method. A method was worked out to apply efficient calculation techniques to the analysis files, which would calculate statistically reliable estimates for risks, including estimates for associated sound deviations. Also considered in doing so was the possibility of being able to arrive at these estimates adequately by using modules from the available software. It was important to inventory only those variables which would make a definite contribution to the interpretation of differences between road stretches or intersections. road stretches or intersections. Determining the class definitions to be distinguished during the inventory also had to take place as efficiently as possible. The developed calculation method was used to analyse the available existing analysis files containing data collected approximately ten years previously. A calculation method that could be elaborated into an interactive analysis method was worked out.By using this method, researchers or policy-making officials with specific questions in mind could determine whether a certain road characteristic would yield statistical differences in risk and in which classes these differences would turn up. It proved possible to draw up a prototype for acquiring experience. See also C 7633 - 7536.
Samenvatting