The SMAA-MABAC approach for healthcare supplier selection in belief distribution environment with uncertainties

Abstract

Supplier selection is critical to the healthcare industry, especially in emergencies. Accurate and timely supply of healthcare resources can reduce economic cost and ensure the safety of people’s lives. However, the healthcare supplier market is enormous and complex; making accurate evaluations of different suppliers and helping managers choose the most suitable one when considering many factors is becoming an important task. Many multicriteria decision-making (MCDM) methods have been applied to supplier selection in existing literature. However, the research on healthcare supplier selection still needs to be completed. In this paper, to deal with healthcare supplier selection problems when there is no sufficient time and information to decide with uncertainty, the stochastic multicriteria acceptability analysis (SMAA) method and multi-attributive border approximation area comparison (MABAC) method are first combined by taking both the advantages of SMAA and MABAC methods when the assessments are expressed by belief distributions (BDs). In the SMAA-MABAC method, the global ignorance in BDs and the uncertain criteria weights are considered by judging the feasible space to derive the final rank of healthcare suppliers with the most possible. The more straightforward calculation process of SMAA-MABAC makes it faster to make an accurate choice among various suppliers. Moreover, it is flexible to generate different recommended suggestions when diverse preference information is added in SMAA-MABAC. Finally, a case study of healthcare supplier selection is examined for managers based on distributed reviews, and targeted selection advice is given concerning different demands. Besides, the sensitive analysis is provided to better understand the influence of the criteria weights on the proposed model. The results of the comparative analysis also show the effectiveness and rationality of the proposed method.