Benchmarking road safety of Montenegro using envelopment analysis

This paper was mainly directed at studying the relevance of benchmarking road safety at national level when only basic information is given. It aims at contributing to this issue by proposing a computational model based on data envelopment analysis (DEA). The three inputs, population size, number of registered vehicles and length of road network, and five outputs which are defined as the ratio of road safety outcomes and some measure of exposure, were used for calculating relative efficiency of 21 municipalities in Montenegro. The period of 2012-2014 was considered due to data availability. DEA looks for the optimum combination of inputs' and outputs' weights for each decision-making unit (municipality) in order to obtain the best possible road safety score based on which a ranking of states can be made. We used output-oriented DEA model with constant return to scale and to rank order the efficient municipalities the cross efficiency rank scale was used. Effective factors for whole analysis are determined by averaging the shares for all 21 DMUs. Also, coloured map is used in order to present the geographical position distribution of the final classifications of the municipalities' relative efficiency. Also, for each municipality that performs relatively poor, a particular municipality will be assigned as a useful benchmark.