Considering the circular economy for designing closed-loop supply chain under hybrid uncertainty: A robust scenario-based possibilistic-stochastic programming

Abstract

Due to the significance of environmental and economic topics and limited resources, integrating Circular Economy (CE) principles is necessary for Supply Chain (SC) to improve sustainable competitive advantage. The integration of CE in SCs leads to a Closed-Loop Supply Chain Network Design (CLSCND) and a circular SC to achieve the economic, environmental, and social aspects of outputs and processes. On the other hand, the integration of CE in CLSCND faces hybrid uncertainties in different time horizons and scenarios due to the strategic and long-term nature of decisions. Due to the impact of random and cognitive uncertainties in the long term, it is necessary to consider these factors in the integration of CE in CLSCND. So, the aim of the present study is CLSCND to achieve the principles of CE and provide a robust scenario-based possibilistic-stochastic programming approach to consider cognitive and random uncertainties simultaneously. The contributions and innovations of the present study are the integration of CE in CLSCND, considering cognitive and random uncertainties at the same time, developing the Me criterion to achieve flexible solutions based on the convex combination of opinions of experts, and proposing the use of the absolute possible deviation to consider the possibilistic deviation. A case study was investigated for CLSCND in the paper industry to assess the presented approach, and the results indicated the accuracy and robustness of the solutions. The numerical simulation results demonstrated the appropriate performance of the proposed method, that the lowest average and standard deviation values of the constraints violation were 601 and 48, respectively, in the developed approach. Analytical findings show that implementing CE in the paper CLSCND provides insights for managers in demand and capacity constraint violations based on different risk levels. Also, this study offers a comprehensive framework for presenting robustness and flexible solutions in CLSC problems based on changes in robustness coefficients and sensitivity analysis of parameters of the CLSCND, which can create a trade-off between CE objectives in uncertain parameters.