Composite performance indicators for helmet and seat-belt enforcement as imprecise data.

Composite road safety performance indicators (RSPIs) are useful tools in regional road safety planning. Among the indicators and data calculated by the World Health Organization (WHO), information on the effectiveness of law enforcement on various risk factors for road casualties were provided, which can be considered as qualitative indicators. The purpose of this study is to analyze the performance indicators related to the percentage of helmet and seat-belt use versus the qualitative enforcement scores attributed by WHO. This analysis was performed for 30 member states of WHO and will show how and with what degree of efficiency the qualitative output of the enforcement score acts versus the input percentage of seat-belt and helmet use. The qualitative nature of the output index has led us to depart the traditional analysis of crisp numerical indicators related to road safety performance and to consider data as imprecise or fuzzy indices. In this study we used two methods including imprecise DEA-based CIs and fuzzy DEA-based CIs, respectively. Results show that the clear score achieved by the Imprecise DEA-based CI model is easy to interpret and use. Whereas, in the Fuzzy DEA-based CI model, the fuzzy indicator scores obtained based on the level of several probabilities are strong in capturing the uncertainties related to human behavior. Both RSPIs are applicable with slight differences that were in the order of countries and the ease of reading the results. We also found that each method has different strengths and that the FDEA-based CIs method is more accurate and more in line with the inputs than the IDEA-based CIs method.