Design and Implementation of an Intelligent Decision Analysis System for E-Sourcing

摘要

Sourcing addresses critical decisions of what to buy, how much to buy, whom to buy from, and how to manage relationship with suppliers. Decision making in sourcing can involve a few hundred offerings each of which is described by several dozen attributes. When using traditional decision analysis techniques, sourcing specialists are often having difficulties in assigning appropriate weights to attributes and feel uncomfortable with decided results. In this paper, we present an innovative approach, where the decision makers only provide ordinal rankings over subsets of offerings. From the information implied by these ordinal rankings, the system derives a set of weights and an overall ranking of all the given offerings. With additional information from the decision maker, these results are iteratively refined. The paper describes the basic concepts and algorithms used as well as the implementation.