Modelling and optimisation of two-sided disassembly line balancing problem with human–robot interaction constraints

Abstract

The mature application of human–robot interaction technology provides a new operation method for remanufacturing waste products, especially in the disassembly process, which further exerts the respective advantages of humans and robots. Therefore, this paper proposes an innovative two-sided disassembly line balancing problem with human–robot interaction constraints (TDLBP-HRI). First, a mixed integer programming model is established for TDLBP-HRI to minimise the number of workstations and smoothness index while simultaneously optimising robot energy consumption and disassembly costs. Then, a discrete squirrel search algorithm (DSSA) is developed to efficiently solve large-scale problems caused by the NP-hard property of TDLBP-HRI. Since SSA is an algorithm that can only solve single-objective linear problems, this research attempts to discretise and improve it to solve multi-objective TDLBP-HRI. In addition, according to the characteristics of the problem, a two-layer decoding strategy is constructed to ensure the uniqueness of the solution, and a fast non-dominated sort is introduced to enhance the multi-objective optimisation capability of the DSSA. To verify the performance of the proposed DSSA, the benchmarks of two other DLBPs types are calculated, and the obtained results are found to be significantly better than those reported in the literature. The correctness of the MIP model and DSSA is verified by comparing the results of the GUROBI solver and DSSA in solving two small-scale TDLBP-HRI examples. Following this, DSSA is shown to outperform four other classical algorithms through different scales cases of TDLBP-HRI. Finally, the proposed model and algorithm are applied to a waste power battery disassembly example. The results show that the proposed model and method can improve the performance of the disassembly line and provide new insights into the application of human–robot interaction technology in the disassembly line.