Road safety performance evaluation based on a multiple layer data envelopment analysis model

Abstract

Currently, performance indicators in the field of road safety are rapidly developed and increasingly used. They are proposed as supportive tools in comparing the level of road safety between various countries and are useful in terms of monitoring changes over time. Moreover, the exploration of a comprehensive composite road safety index which is a combination of individual indicators is attractive and desirable due to its political importance and operational relevance in decision making. Here, we start from four types of indicators as defined in the SUNflowerNext report (2008). A data set consisting of 21 indicators for 26 European countries is analyzed, and data envelopment analysis (DEA) as a performance measurement technique is applied to combine these 21 indicators into a composite index. In particular, the concept of hierarchical structure is embodied in the model thereby giving a more detailed insight into the layered hierarchy of the indicators. In addition, the presence of both quantitative and qualitative indicators is taken into account. In the end, a separate, best possible model is constructed for each country, the most optimal road safety index score is computed, and the weights assigned to each layer of the hierarchy are analyzed.