Multi-objective optimization of buffer allocation for remanufacturing system based on TS-NSGAII hybrid algorithm

Remanufacturing is of great importance for environmental protection and sustainable development, while the uncertainty in returns' quality has brought huge challenge for the design and operation of remanufacturing systems. By considering returns' quality, this study is to optimize the buffer allocation with maximum throughput rate and minimum work in process (WIP) concurrently. Decomposition-extension-Markov approach is adopted to establish the model and obtain the performance of the system. A novel tabu search non-dominated sorting genetic algorithm-II (TS-NSGA II) is put forward to search the optimal solution, and the Pareto-optimal solutions are obtained. A case study is provided to demonstrate the effectiveness of the proposed approaches. The main findings of the study are as follows: (1) Compared with the previous studies, a Pareto optimization can maintain the diversity of the solutions, thus it is favorable to make better decisions for multi-objective buffer allocation. (2) TS-NSGA II can obtain optimal solutions closely enough to the Pareto frontier, and it has significant advantages in convergence, diversity and running time. (3) Buffer capacity and its allocation have important effect on the performance of remanufacturing system. For WIP, the buffer capacity is the most critical influence factor; for the throughput ratio and discarded ratio, buffer capacity is the secondary factor just behind the process route. The above achievements provide an valuable reference for the optimal design of remanufacturing system.