A bi-level decision-making system to optimize a robust-resilient-sustainable aggregate production planning problem

• Published in: Expert systems with applications Vol. 228; p. 120476
• Main Authors: Tirkolaee, Erfan Babaee, Aydin, Nadi Serhan, Mahdavi, Iraj
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.10.2023
• Subjects: Bi-level decision-making system; BWM-WASPAS-Neutrosophic; Resilient supplier selection; Robust optimization; Sustainable aggregate production planning; Workforce productivity; Resilient supplier selection; Sustainable aggregate production planning; Workforce productivity; Robust optimization; BWM-WASPAS-Neutrosophic; Bi-level decision-making system
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2023.120476

This study introduces a sustainable-robust aggregate production planning (APP) problem taking into account workforce productivity, outsourcing option and supplier resilience. In this regard, a bi-level decision-making system is designed using multi-attribute decision-making (MADM) and multi-objective decision-making (MODM) models, respectively. In the MADM section, a hybrid method called BWM-WASPAS-Neutrosophic based on best-worst method (BWM) and weighted aggregated sum-product assessment (WASPAS) under type-2 neutrosophic number (T2NN) is utilized to investigate resilient supplier selection. To design the MODM section, a multi-objective mixed-integer linear programming (MILP) model is suggested which is then treated with the help of weighted goal programming (WGP) method. The multi-objective model honors sustainable development goals as it aims not only to minimize total cost and maximize weighted purchasing from suppliers, but also to minimize negative environmental impacts simultaneously. Robust optimization (RO) technique is then implemented to the MODM model to address the demand uncertainty. In order to validate the suggested methodology, a real case study from the literature is examined. The obtained results reveal the efficiency of our decision-making system in finding the optimal policy in less than 1 s where sensitivity analyses also contribute to practical managerial implications. Finally, it is revealed that the total weighted purchase from suppliers has the highest sensitivity to the conservatism levels, which determine the extent to which our uncertain demand parameter can deviate from its nominal value.