Advancements of Multi-Criteria Decision-Making Approaches Integrated with Artificial Intelligence and Machine Learning in Sustainable Supply Chain Management

Abstract

This work examines the application of multi-criteria decision-making (MCDM) techniques in conjunction with artificial intelligence (AI) and machine learning (ML) within sustainable supply chain management (SSCM). Traditional MCDM methods struggle with large datasets and decision-making environments that change dynamically. AI and ML help get around these problems by making it easier to deal with uncertainty and by automating some parts of the analysis. The review discusses new methods, such as Measurement of Alternatives and Ranking according to Compromise Solution (MARCOS), Simple Weight Calculation (SIWEC), and the Aczel Alsina Weighted Assessment (ALWAS). These methods give decision-makers more options for how to look at environmental, social, and economic factors. AI and ML tools can help MCDM methods work better with dynamic data, which improves risk assessment and resource allocation. Combining such methods makes MCDM approaches better able to deal with complicated decision-making situations and gives a more complete picture of how well a sustainable supply chain is doing. The review also offers practical guidance on selecting suitable learning algorithms according to the available data and their intended role in the MCDM process. In addition, it distinguishes between reporting gaps, which can be resolved through more consistent reporting practices, and conceptual gaps, which require the development of new methodological approaches.