Requirement-driven sustainable supplier selection: Creating an integrated perspective with stakeholders' interests and the wisdom of expert crowds [1]

Driven by the pressure from various stakeholders to embrace sustainability and from the market to enhance competitiveness, more and more companies have been committed to conducting sustainable supplier selection (SSS). SSS is a requirement-driven multi-criteria decision-making (MCDM) problem and the existing pieces of literature have attempted to integrate quality function deployment (QFD) with various MCDM methods to model the problem. However, these models have not conducted SSS based on stakeholders’ satisfaction from alternatives and used the weights of criteria obtained from stakeholders’ preferences for requirements, which is used to model the compensation among satisfaction levels of different requirements, to guide the compensation among alternatives’ performances on different criteria. In order to address these deficiencies, this paper develops a novel generalized (QFD)-based MCDM structure for SSS, based on which we further design a QFD-based multi-stakeholders and multi-experts MCDM model considering stakeholders’ consensus of interest and wisdom of expert crowds. Firstly, multiple stakeholder groups with multiple participants in each one are asked to output their expectations and preferences for their own requirements; within each group, the algorithm for consensus reaching process (CRP) is designed to formulate the acceptable collective expectations, and the best-worst method (BWM) method integrated with an algorithm for consistency improving process (CIP) and an algorithm for CRP is developed to weight requirements. Secondly, multiple experts use basic uncertain linguistic information (BULI) to characterize performances of alternatives on criteria and the relationship between requirements and criteria, and a method of weighting individual judgments with the capability of drawing on the wisdom of crowds (WOC) while accommodating organizers’ trust level in reliability given by experts is used to facilitate formulating a collective judgment. Thirdly, for each alternative and each requirement of each stakeholder group, we can quantify the comprehensive performance of the alternative on the requirement with the help of QFD, based on which and the group’s expectation we can use the value function of prospect theory to quantify the group’s satisfaction of the requirement from the alternative; aggregating the weighted satisfaction levels of requirements gives rise to the comprehensive satisfaction level of each group from each alternative, and then the Maximin decision rule is introduced to rank alternatives because of its ability to accommodate the non-compensation among different stakeholder groups. Finally, a case study is conducted to investigate the validity and effectiveness of the proposed model.