Scholarlink

<p>Supply chain management involves the planning, directing, and controlling of material flows from raw material suppliers to end consumers. Within this context, supplier selection represents a critical decision-making process, as inappropriate supplier choices can adversely affect organizational efficiency, cost-effectiveness, and long-term competitiveness. The primary objectives of supplier selection include minimizing procurement risk, maximizing overall buyer value, and establishing sustainable, long-term buyer–supplier relationships. This study presents a systematic review of 110 reputed journals articles published between 2009 and 2025, collected from major scientific databases including Scopus, Web of Science, Elsevier, Science Direct, Springer, and IEEE Xplore. The objective of this review is to examine the current state of research on the application of artificial intelligence (AI) in supplier selection, identify dominant methodological trends, and highlight key challenges and research gaps. The findings reveal that while traditional multi-criteria decision-making (MCDM) methods provide structured and transparent decision support, they face limitations in handling large-scale, dynamic, and non-linear supplier data. In contrast, AI-based models demonstrate superior capability in processing complex datasets and capturing non-linear relationships. Consequently, hybrid AI–MCDM approaches, which integrate expert judgment with data-driven learning, emerge as the most robust and reliable solutions for modern supplier evaluation. The review identifies artificial neural networks, tree-based ensemble models, logistic regression, and metaheuristic algorithms as the most frequently applied AI techniques. However, challenges related to data quality, model interpretability, computational complexity, limited real-world validation, and insufficient integration of expert knowledge continue to hinder practical adoption. Unlike existing reviews that focus on isolated methodologies, this study provides a structured and comparative synthesis of MCDM, AI-based, and hybrid models, offering practical guidance for the development of scalable and interpretable supplier decision-support systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Systematic Review of Artificial Intelligence Revolution in Supplier Selection for Multi-criteria Decision-Making

  • 期刊论文

  • 发表日期：

    2026年4月18日

查看全文
  • Santosh Kr. Gupta,
  • Shikha Chadha,
  • Anubhava Srivastava

摘要

Supply chain management involves the planning, directing, and controlling of material flows from raw material suppliers to end consumers. Within this context, supplier selection represents a critical decision-making process, as inappropriate supplier choices can adversely affect organizational efficiency, cost-effectiveness, and long-term competitiveness. The primary objectives of supplier selection include minimizing procurement risk, maximizing overall buyer value, and establishing sustainable, long-term buyer–supplier relationships. This study presents a systematic review of 110 reputed journals articles published between 2009 and 2025, collected from major scientific databases including Scopus, Web of Science, Elsevier, Science Direct, Springer, and IEEE Xplore. The objective of this review is to examine the current state of research on the application of artificial intelligence (AI) in supplier selection, identify dominant methodological trends, and highlight key challenges and research gaps. The findings reveal that while traditional multi-criteria decision-making (MCDM) methods provide structured and transparent decision support, they face limitations in handling large-scale, dynamic, and non-linear supplier data. In contrast, AI-based models demonstrate superior capability in processing complex datasets and capturing non-linear relationships. Consequently, hybrid AI–MCDM approaches, which integrate expert judgment with data-driven learning, emerge as the most robust and reliable solutions for modern supplier evaluation. The review identifies artificial neural networks, tree-based ensemble models, logistic regression, and metaheuristic algorithms as the most frequently applied AI techniques. However, challenges related to data quality, model interpretability, computational complexity, limited real-world validation, and insufficient integration of expert knowledge continue to hinder practical adoption. Unlike existing reviews that focus on isolated methodologies, this study provides a structured and comparative synthesis of MCDM, AI-based, and hybrid models, offering practical guidance for the development of scalable and interpretable supplier decision-support systems.