Determining production batch size an‎d production sequence is critically important in supply chain management due to its direct impact on cost, inventory, an‎d productivity. In multi-product, multi-stage systems such as flow shop models, precise planning is essential. One of the main challenges in this area is deman‎d uncertainty, which has been managed using robust optimization models to provide resilient an‎d sustainable solutions. Environmental issues such as energy consumption an‎d emissions have also been incorporated into the model to ensure more sustainable decision-making alongside cost optimization. The proposed model, which involves mixed-integer programming including decision variables for production quantity, sequencing, resource allocation, an‎d inventory management over multiple periods, is solved in small-scale instances using GAMS software an‎d the CPLEX solver due to the NP-hard nature of the problem. For larger instances, the Gray Wolf Optimizer (GWO) metaheuristic algorithm is employed. To eva‎luate the quality an‎d accuracy of the algorithm’s results, five different methods for calculating the lower bound of the problem were used, ensuring that the solutions are close to optimal. The results show that the algorithm achieved optimal solutions in all small instances an‎d performed satisfactorily in other instances with a gap of less than 1%. For medium-sized problems, the algorithm achieved an average error of 1.21% with a runtime of 33 seconds, while for large problems, the average error was approximately 3.21% with a runtime of 470 seconds, making it suitable for practical applications. This study demonstrates that the combination of robust optimization an‎d the Gray Wolf Optimizer provides an effective an‎d efficient approach for solving complex production problems under uncertainty an‎d environmental constraints, with broad applicability across various industries facing similar challenges.

مهدي بيجاري

علي شاهنده نوك آبادي , فرشته پرورش