Skip to main content

Multilayered safety objectives


The purpose of this project is to determine the traffic situations and circumstances in which RTI solutions to accidents are to be of greatest likely benefit. These analyses will be provided to other DRIVE projects concerned with the design and evaluation of RTI systems.
Safety objectives for road transport informatics (RTI) development and implementation have been formulated as follows:
traffic accident statistics from some European countries were analysed, and those traffic safety problems with the highest costs were defined;
the most serious (ie costly) traffic safety problems were analysed using a behavioural interactional frame of reference, and RTI measures were defined that may decrease the number and severity of accidents of the given category (as far as possible, the system effects of these RTI measures were also considered);
RTI measures planned, designed or implemented within DRIVE or PROMETHEUS projects were analysed from the perspective of their possible and expected traffic safety effects;
other DRIVE projects were surveyed and their objectives analysed by defining their safety and nonsafety objectives.

The main safety problems identified for which RTI measures have been suggested are:
single vehicle accidents on motorways;
rear end accidents on motorways;
single vehicle accidents on rural roads;
head on collisions on rural roads;
crossing collisions on rural roads;
crossing collisions on urban intersections;
pedestrian and bicyclist accidents in urban areas.
The project, which is a continuation of the work done in V1040, requires the ordering of objectives in the estimate of the likely benefits from specific RTI systems. Our initial approach will be to estimate the following quantities whenever possible and as accurately as possible:

- Na (the number of accidents in a given category)
- Ca (the average cost of accidents in that category)
- Av (the avoidability of the accident in purely technical terms)
- Rc (risk compensation in human reactions to the RTI system)

These four quantities would be expected to combine in a multiplicative manner so as to predict the expected benefit of a particular measure.

Expected safety benefit = Na.Ca.Av.Rc (1).

Equation (1) is spuriously exact, since there will always be considerable uncertainty in the estimate of any of the quantities in it.

Its value is in indicating the multiplicative relationship between them. To estimate a likely benefit, each of the four quantities must be estimated with comparable accuracy. However it is possible to compensate for some of the inaccuracies by using more than one method to estimate the quantities.

The central part of the project is a statistical analysis of the accident data, to determine the complex interactions of variables, and to determine a model of the interaction of driver, environment and vehicle factors.

This model will be used to make a preliminary identification of situations in which RTI interventions are likely to be effective. These judgements will be influenced by the expertise of the team in human factors, including risk compensation and complex systems analysis. Consultations are being held with other DRIVE projects to adapt the model to their particular needs. In doing so the project shall also become aware of opportunities for RTI interventions (e.g. collision warning, and avoidance devices) which are not yet being considered.

Key Issues

Analysis of accident statistics to identify high-risk conjunctions of variables;
Modeling the interaction between driver, environment and vehicle;
Identification of situations in which RTI interventions are likely to be effective.


University of Lund

Participants (3)

Communication and Management Systems Unit
Loughborough University of Technology
United Kingdom
The Elms Elm Grove
LE11 1RG Loughborough
Technische Universitaet Muenchen
Lichtenbergstrasse 4
85748 Garching