Remanufacturing planning under uncertainty

The circular economy concept involves a systematic approach towards sustainable development by increasing the productivity of resources and the creation of business opportunities. Remanufacturing is recognized as one of the most profitable and environmentally conscious options of the circular economy. Remanufacturing emphasizes on value-added recovery rather than just material recovery. A remanufacturing process chain includes disassembly, cleaning, inspection, reconditioning and reassembly stages to recover the functionality and value of returned products. However, the profitability of remanufacturing is significantly affected by the variability of EOL(End Of Life) product conditions. Among different steps of the remanufacturing process chain, disassembly is highly affected by the negative impact of the EOL products variability. The effect is the highly uncertain disassembly tasks times. This uncertainty can significantly disturb performance and feasibility of remanufacturing since disassembly is the prerequisite of other steps. Consequently, this leads to a lack of robustness in disassembly lines designed without considering these challenges. Therefore, providing the efficient disassembly systems considering these challenges is the requirement of having profitable remanufacturing system.\par This work develops a mathematical optimization model to find the optimal design of disassembly lines with the objective of profit maximization. This model jointly optimizes the disassembly tasks sequencing, the allocation of tasks to the workstations and the capacity of buffers under uncertainty of tasks times in order to achieve the desired cycle time and service level. This is the first work in the disassembly line context that interactively solves the most important design aspects of the disassembly lines, where the inter-departure time analysis of finished components for evaluation of meeting the desired cycle time is proposed. The mathematical optimization model is proposed in two versions. In the first version, the sequence of disassembly tasks, assignment of workstations and buffer allocation are optimized where the expected value of inter-departure time meets the desired cycle time. On the other hand, in the second version, the same problem is optimized in which the service level of the line in terms of the probability of meeting desired cycle time by the inter-departure time of the components from the line is satisfied. The optimization model is solved by dividing the whole problem into two sub-problems: 1) disassembly sequencing and workstations assignment, and 2) buffer allocation. Then, the two sub-problems are solved independently and are jointly optimized by an iterative algorithm in order to achieve the optimal/near optimal disassembly line design. Besides, an Extensive Search Method (ESM) is developed to validate the accuracy of the solution achieved by the two-step method. Also, the high accuracy of the two-step method is validated by several experiments.\par The main achievement of the proposed optimization model is providing the profitable disassembly line design which correctly satisfies the desired cycle time in comparison to the traditional line designs in which mostly the desired cycle time is not achieved due to neglecting the inter-departure time between finished product as the right measure to evaluate the achievement of cycle time. It is shown that in order to reach the higher service level, the more number of buffers is required, and in some cases, the design of the line is required to be modified. The benefits of the proposed model are as well validated within a real case study dedicated to the remanufacturing of mechatronic components in the Knorr-Bremse company. The achieved results show the successful implementation of the proposed method in the remanufacturing section of Knorr-Bremse, and the numerical results prove that the proposed method can improve the efficiency and robustness of the disassembly line which in turn supporting remanufacturing.