Normalization Techniques' Effect on Multi-Criteria Methods for Decision Making

Abstract

This work aims to add to the literature by contrasting the impact of normalization techniques on deciding upon a suitable Web service using the AHP-VIKOR technique. The weights of each criterion are computed using AHP and the ranking of the web services, according to a user preferred criterion, is obtained using the VIKOR method. One of the most significant steps in solving multi criteria decision-making (MCDM) problems is the normalization of the decision matrix. The consideration for the normalization of the data in a judgment matrix is an essential step as it can influence the ranking list. This study also investigates the effects of normalization on an AHP-VIKOR hybrid method in the selection of Web services. The Web services taken into consideration in this study provide comparable functionalities, but with varying levels of service quality (QoS). Ten web services were chosen for the study's purposes. To select the most appropriate Web services, each of these services was assessed against five different QoS requirements: response time, throughput rate, latency, availability, and reliability. The five normalization techniques employed are linear sum, linear max, linear maxmin, enhanced accuracy and vector normalization. It was observed that different ranking lists are produced when applying different normalization techniques to the VIKOR method and normalization has an influence on the final ranking list. In future studies, a comparative analysis of different normalization techniques on other MCDM approaches are proposed.