Author(s)

S. Basbas

Abstract

The development of mathematical models in order to relate the number of accidents with the geometrical and functional characteristics of road networks is always an important issue in road safety. The first attempts started in the early 50s and since then a lot of research has been made towards this direction. As a result a number of tools have been developed worldwide. The Transport Research Laboratory (TRL) developed the SafeNET software package, which allows estimates to be made of the number of personal injury accidents per year for particular elements of a road network. Within the framework of this paper the results from the use of the SafeNET software package in the central area of the city of Thessaloniki are presented. The network consists of four links and four nodes. SafeNET has proved to be a very useful tool to assist the work of traffic engineers. Keywords: road accident, risk models, SafeNET. 1 Introduction The problem of how to select junctions or road sections in order to take measures for the improvement of road safety level has to do with the accident data availability, and more especially with the knowledge of the accidents, which would occur in the case where no measures are implemented. One way to deal with this problem is to take into account for the estimations the historical data (e.g., road accidents which took place during the last 3 to 5 years) but there is always the question of reliability due to regression-to-mean (RTM), due to trend effects etc. The use of Empirical Bayes (EB) approach appears to provide very

Keywords

road accident, risk models, SafeNET.