Multi-criteria approach to stochastic and fuzzy uncertainty in the selection of electric vehicles with high social acceptance

Abstract

In recent years, a number of legal solutions have been adopted in Poland aimed at reaching the number of 1 million electric cars in use in 2025. Therefore, an important research issue is the assessment of electric vehicles available on the Polish market and the identification of vehicles that meet consumers' expectations to the greatest extent. The assessment of electric vehicles is a multi-criteria issue, characterized by a number of uncertainties related to their operational parameters, as well as the preferences of individual users and, more broadly, the preferences of society. A novelty of the article is the application of the fuzzy multi-criteria decision aid method called NEAT F-PROMETHEE (New Easy Approach To Fuzzy PROMETHEE), combined with the Monte Carlo method and elements of the SMAA (Stochastic Multicriteria Acceptability Analysis) method for the assessment of vehicles under uncertainty conditions. The theoretical contribution of this research therefore includes the synthesis of a fuzzy and stochastic approach to decision-making support, supported by outranking and incomparability relationships. Such a set of approaches to uncertainty makes it possible to improve the accuracy of decision-making, because the approaches indicated verify each other's results. Moreover, the application of the NEAT F-PROMETHEE method has solved the problem of evaluating electric vehicles from the perspective of a single decision-maker, while the combination of NEAT F-PROMETHEE and the stochastic approach has made it possible to simulate preferences of the society. Although there are many uncertainties in the decision-making problem, the approach has allowed to identify almost unambiguously the electric vehicle that is likely to gain the highest acceptance. As a result of the conducted research it was found that the approaches to uncertainty based on fuzzy sets, outranking relations and stochastic analysis complement each other, allowing the decision-maker to conduct a wider analysis of the imprecision of the obtained solution.