Induced Ordered Weighted Evidential Reasoning Approach for Multiple Attribute Decision Analysis with Uncertainty

Abstract

We are primarily concerned with the problem of aggregating multiple attributes with uncertainty to form an overall decision function. We introduce a new type of approach for aggregation called an induced ordered weighted evidential reasoning IOWER approach, which is inspired by an induced ordered weighted averaging operator and the evidential reasoning ER approach. In the IOWER approach, we use a belief decision matrix combined with an induced ordered weighting vector for problem modeling and the Dempster-Shafer theory of evidence for attribute aggregation. It is proved that the original ER algorithm is a special case of the IOWER algorithm. Then we examine the properties of the IOWER approach. One key point in the IOWER approach is to reorder the arguments in the form of distributed assessment structure. A kind expected utility order-inducing variable is proposed in the IOWER approach, which can make the alternative's advantages prominent. Finally, we present an illustrative example in which the result obtained with the new aggregation approach can be seen.