Urban development and road safety.

There is still not sufficient in-depth knowledge available concerning the type and extent of the interdependencies between the large number of factors that influence, on the one hand, urban and transport planning, and on the other hand, road safety. If such in-depth knowledge did, in fact, exist this information would make it possible to recommend courses of action to urban planning authorities. The present study aims to remove this data deficit so that planning authorities may be able to assess not only the effects expected from individual measures but also to evaluate, with respect to the level of road safety, an entire planning area. The empirical study is based on 21 typical residential areas for which a total of 56 characteristics (variables) relating to the following sectors have been identified as descriptors of the accident situation: transport; local development structure; infrastructure facilities other than those intended for transport; and population structure. The statistical analysis was undertaken in two steps: correlation analysis of all variables; identification of the determinants of significance for road safety; and establishment of a non-linear regression model based on seven variables to express the relationship between transport provisions and traffic volumes as well as the quality of supply and demand in quantitative terms. By means of the regression model the accident load can, in principle, be assessed for any given residential area. The correlation analysis showed that there are still further influential factors which are of importance for road safety. Attempts were therefore made to identify the variables displaying a significant relationship to the model variables and which could thus, owing to their effects on the latter and the resulting linkage with road safety, be considered as "background variables." Whereas the majority of the model variables are to be ascribed to the transport sector, the so-called "background variables" mainly relate to other areas of urban development. Based on the elasticities of the variables describing various levels of dependencies, an evaluation procedure has been developed. Based on this the applicability of the research findings to urban planning operations can be assessed. The contributions to road safety of the different variables can be quantified and compared, and recommended courses of action can be derived from this procedure.(ABSTRACT TRUNCATED AT 400 WORDS)