Disaggregated accident models.
C 19375 (In: C 19360) /82 /91 / ITRD E110190
Roine, M. & Kulmala, R.
In: Working together for a better future : proceedings of the 26th International Symposium on Automotive Technology and Automation (ISATA) dedicated conference on road and vehicle safety, Aachen, Germany, 13th-17th September 1993, p. 357-363, 10 ref.
|Samenvatting||Accident prediction models are useful tools in road safety work. The authors have been developing models for rural road sections and junctions during the last years. These models have been integrated into road planning systems estimating the effects of road improvements on road safety and selecting accident black-spots for further analysis. The basic idea of these models is to predict the expected lumber of injury accidents in various locations. The authors model accidents as the product of accident risk and exposure. The authors have also applied The Empirical Bayes approach to improve the estimates of the expected number of accidents. Accident prediction models explaining the variation of accidents as a function of risk factors are usually aggregated models. These models have been popular because safety problems are usually measured with the number of accidents. And aggregate models explain the expected number of accidents. The relationship between accident occurrence and risk factors can be obscured in aggregated models. These models may be unable to reveal the causal relations behind the accident process, although the models may be adequate and fit well with the data. It is also difficult to integrate the variation of road user behaviour into aggregated models. The practical problems why there is interest in disaggregate models are related to winter traffic safety. The authors are studying the effects of tyres and their condition on road safety. We are especially interested in what would be the effects on road safety if drivers would replace studded tyres with friction tyres in winter time. This problem is very complicated and can not be studied with aggregated models hiding the large variation of driver characteristics and road-user behaviour. (A)|
|Full-text||Beschikbaar Niet beschikbaar, klik om contact op te nemen voor een digitalisatie verzoek|
|Suggestie?||Neem contact op met de SWOV bibliotheek voor uw opmerkingen|