Citation

Tanzi TJ, Guiol R, Laurini R, Servigne S. Safety Sci. 1998; 30(1-2): 9-23.

Copyright

(Copyright © 1998, Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Mobility and flexibility are decisive elements in our modern industrial society. Every day more and more people and goods have to be transported rapidly, at a lower cost and, above all, safely. Due to continuous economic growth, the volume of traffic increases daily; the result is a gradual saturation of road and motorway networks. Whatever the case, the fundamental requirement remains the same: the increase in traffic (Transport, 1995. Le transport intelligent: gestion du temps et securite des deplacements. Actes du Colloque Ministeres de l'Equipement, des Transports et de Tourisme, Ecole Nationale de Ponts et Chausees) must not take place at the expense of safety. In view of this fact, safety improvements are attained by means of the implementation of active methods: these are primarily intelligent systems (Guiol, 1994. Neutralisation automatique de voie sur voies rapides et autoroutes. Associations des Ingenieurs et Anciens Eleves de l'ENPC) dealing with traffic management whose purpose is to help control the flow of vehicles according to the situation on the basis of information processed as it appears and to avoid as many secondary accidents as possible (accidents which are the consequence of a first accident). To complement these assistance systems, we set up non-contact measuring systems, allowing us to obtain a preventive vision of an accident-provoking situation, depending on the conditions at a given time. We obtain these results at any point throughout the network, simply by using a geometric description of traffic conditions (flow, speed, taking into account any incidents as they happen) and meteorological conditions. The purpose of our survey consists of setting up an appropriate algorithmic system capable of supplying various data and syntheses as they occur, to operators in charge of managing the motorway to help them make decisions that are required in the course of their work. To begin with, we used traditional methods of data processing in order to obtain a clear comprehension of the phenomena under review. Then, based on surveys already carried out, we moved into a phase of modelling data and processes. We established the various relations between traffic data and those concerning the geometric description of the road. These characteristics allowed us to define the parameters which influence these phenomena and hence better understand traffic problems.