Scholarlink

<p>The agricultural supply chain faces increasing challenges due to environmental degradation, resource constraints, and market uncertainty, emphasizing the need for sustainable and resilient supply chain design. In response, this study proposes a comprehensive multi-stage data-driven decision framework for configuring a viable agricultural supply chain network that integrates circular economy principles. The methodology combines multi-criteria decision-making (MCDM), optimization, and machine learning techniques to enhance decision quality under uncertainty. Initially, potential retailers are evaluated based on viability and circular economy indicators using an integrated fuzzy best–worst method (FBWM) and CatBoost Regressor. A multi-objective mathematical model is then developed to design the supply chain, and to address parameter uncertainty, a hybrid robust possibilistic optimization method informed by machine learning is applied. The final model is solved using Lexicographic Chebyshev Revised Multi-Choice Goal Programming (LCRMCGP). Overall, the main contribution of this research is to develop an effective data-driven model to incorporate viability and circular economy metrics into agricultural supply chain network design. The results confirm the applicability, robustness, and effectiveness of the proposed framework. Sensitivity analyses show that increasing capacity improves total profits and reduces environmental impacts, while node criticality. Additionally, higher demand raises total profits, environmental damages, social impacts, and node criticality. Also, implementing a blockchain-based information sharing system significantly improves visibility, albeit with increased overall costs.</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

Configuring a viable agricultural supply chain network based on the circular economy aspects: a multi-stage data-driven decision framework

  • 期刊论文

  • 发表日期：

    2026年7月3日

查看全文
  • Mohammad Vahidi,
  • Ahmad Mehrabian,
  • Hossein Amoozad Khalili

摘要

The agricultural supply chain faces increasing challenges due to environmental degradation, resource constraints, and market uncertainty, emphasizing the need for sustainable and resilient supply chain design. In response, this study proposes a comprehensive multi-stage data-driven decision framework for configuring a viable agricultural supply chain network that integrates circular economy principles. The methodology combines multi-criteria decision-making (MCDM), optimization, and machine learning techniques to enhance decision quality under uncertainty. Initially, potential retailers are evaluated based on viability and circular economy indicators using an integrated fuzzy best–worst method (FBWM) and CatBoost Regressor. A multi-objective mathematical model is then developed to design the supply chain, and to address parameter uncertainty, a hybrid robust possibilistic optimization method informed by machine learning is applied. The final model is solved using Lexicographic Chebyshev Revised Multi-Choice Goal Programming (LCRMCGP). Overall, the main contribution of this research is to develop an effective data-driven model to incorporate viability and circular economy metrics into agricultural supply chain network design. The results confirm the applicability, robustness, and effectiveness of the proposed framework. Sensitivity analyses show that increasing capacity improves total profits and reduces environmental impacts, while node criticality. Additionally, higher demand raises total profits, environmental damages, social impacts, and node criticality. Also, implementing a blockchain-based information sharing system significantly improves visibility, albeit with increased overall costs.