A new integrated approach based on base-criterion and utility additive methods and its application to supplier selection problem

Abstract

Supplier selection is one of the biggest challenges facing manufacturing companies in an increasingly competitive environment. Selecting the right supplier can minimize the cost of raw materials, which constitute a large portion of the final production cost. During the past decade, attention has been drawn to the supplier selection problem, a complex process in which decision-makers assess multiple quantitative and qualitative characteristics. Approaches such as applied individual or hybrid Multi-Criteria Decision-Making (MCDM) methods have also been developed to elucidate this problem. This research aims to offer a novel integrated MCDM method for the supplier selection problem based on the combination of Base Criterion (BC) and Utility Additive (UA) methods. In this approach, one of the criteria is selected as the base criterion. This criterion is compared with other criteria by considering the DM's preferences. The main contribution of this study is to present an integrated approach that simultaneously calculates the criteria weights and the rank of alternatives, leading to fewer pairwise comparisons and more consistency. Also, the robust mathematical foundation of the suggested model, which incorporates utility functions, assures that the results align with decision-maker preferences as much as possible. The Base Criterion Utility Additive (BCUA) model proposed in this paper, formulated as a nonlinear programming (NLP) model, is solved using the Lingo software package. A numerical example and a real case in the electronics industry have been presented to corroborate the applicability and effectiveness of the BCUA model. Sensitivity analysis and a comparative analysis of the proposed method are discussed in comparison to already existing MCDM methods, including the Fully Consistency Method (FUCOM), Ordinal Priority Approach (OPA), Level Based Weight Assignment (LBWA), and Defining Interrelationships Between Ranked Criteria (DIBR). These analyses indicate that the criteria weights and alternative rankings are robust to changing the parameters δ and λ. In addition to selecting the most suitable supplier, the model presented here can also be used in any decision-making problem with multiple attributes and alternatives, particularly when decision-makers are faced with many alternatives.