Application of Correlation Test to Criteria Selection for Multi Criteria Decision Making Problems in Domain of Logistics Systems

Abstract

Generally, at considering any selection problem in domain of logistics and logistics systems there is a great number of technically feasible alternatives, and the task of designer is to choose from the set of possible solutions the one that best meets the technical and economic conditions defined by the terms of reference. Multi criteria decision making (MCDM) models are widely used in selection problems in the literature. Starting base at criteria defining is the fact that at solving each problem we can adopt various number and kind of criteria depending on corresponding decisions and information available. Also, an unique set of criteria of considered problem usually is not available to a decision maker. In this work the correlation test was used for getting the set of independent criteria and reducing their number to operating and acceptable level. It has been shown in the paper that the obtained set of independent criteria still fully represents the characteristics of the considered selection problem.