Comparative Analysis of the MCDM Methods with Multiple Normalization Techniques: Three Hybrid Models Combine MPSI with DNMARCOS, AROMAN, and MACONT Methods

Several multi-criteria decision-making methods have been developed to solve complex decision problems encountered in business and daily life. These methods offer a systematic approach to evaluating multiple decision alternatives and conflicting criteria. The normalization stage in the multi-criteria decision-making process is important in evaluating the contribution of criteria or alternatives to the process in a fair, consistent, comparable, and objective way. Various methods employ one or more normalization techniques, and the combined use of multiple normalization techniques allows for a comprehensive analysis. In this study, Double Normalized Measurement of Alternatives and Ranking According to COmpromise Solution (DNMARCOS), Alternative Ranking Order Method Accounting for Two Step Normalization (AROMAN), and Mixed Aggregation by Comprehensive Normalization Technique (MACONT) methods used multiple normalization techniques are compared and evaluated for a robot vacuum cleaner selection problem. The relations of the ranking results were evaluated by correlation analysis. The performance comparisons of the methods were made in terms of the final scores' standard deviations and the methods' computational complexity. The findings indicate that DNMARCOS has the best performance among the three methods and MACONT has the lowest performance.