A new component depolymerization method for selective assembly optimization of complex products with parallel structures

The advancement of intelligent manufacturing technologies is driving production toward higher precision and efficiency, pushing traditional quality control methods to their limits. Selective assembly enables high-precision products to be assembled from low-precision components, which shows great potential for application in smart assembly lines powered by digital measurement and augmented reality technologies. However, existing selective assembly methods are only applicable to simple linear assemblies and fail to address complex products with numerous tolerance information and intricate assembly connections. Two immature approaches can be used to address the challenges, both of which have significant limitations. One overlooks intricate tolerance information, leading to large deviations, while the other indiscriminately groups all tolerance information, resulting in exponential group proliferation. In this paper, a new component depolymerization method for the selective assembly of complex products is proposed to address the challenges, which integrates the assembly connection separation model, the component precision aggregation model, and the optimization algorithm. The proposed assembly connection separation model comprehensively considers all tolerance information in serial and parallel connections without any omission. The established component precision aggregation model efficiently consolidates the tolerance information of each component, overcoming the problem of exponential growth in group numbers. The genetic algorithm is also developed to provide an excellent solution. Two typical cases with parallel structures are used to demonstrate the superior performance of the component depolymerization method for the selective assembly of complex products compared to traditional methods.