Risk Management Based on Hybridized TOPSIS Method Using Genetic Algorithm

Abstract

Risk management decision-making is essentially based on the choice of context, this context mainly depends on the criteria chosen and the weightings allocated or calculated for each criterion. Traditional “multicriteria decision-making methods” are generally subjective and depend to a large extent on the weightings expressed by the decision-maker. Our work consists of the proposal of a methodology based on the use of genetic algorithms for the calibration of the weighting coefficients necessary in the process of using the multi-criteria decision method TOPSIS. The aim is to automate the risk management decision-making process and to compare the results obtained with genetic algorithms with the results obtained using conventional multi-criteria decision-making methods. Indeed, the results obtained by using the genetic algorithms to the data in the TOPSIS matrix without the usual intervention of the decision maker in the choice of the weighting of the coefficients are satisfactory. Hence the efficiency of our approach in comparison with the conventional TOPSIS method.