Product sequencing in multiple-piece-flow assembly lines

Modern markets demand mass customization, that is, the manufacture of customized products at low cost. Mass customization represents a major challenge for the organization of assembly lines, which were originally designed for the manufacture of homogeneous products. The multiple-piece-flow assembly line is an organizational innovation that can address this challenge. Here, several customized workpieces, each associated with a separate customer order and, hence, a separate due date, are handled simultaneously in one cycle. Consequently, the idle times decrease as do the manufacturing costs. Multiple-piece-flow assembly lines are used, for instance, in manufacturing industrial equipment.To the best of our knowledge, this paper is the first to investigate product sequencing in multiple-piece-flow assembly lines. We formalize the underlying planning problem, establish a mixed-integer model, examine its relation to several classic optimization problems, and describe useful problem properties. We leverage these properties to design an effective iterative variable neighborhood heuristic (IVNH). A detailed simulation based on real-world data and the rolling-horizon planning framework confirms that the IVNH is well suited for practical use. Furthermore, extensive computational experiments on well-structured randomly generated data sets show that the IVNH identifies optimal or near-optimal solutions within short run times. It outperformed an off-the-shelf optimization software, and in certain practice settings, the IVNH was even able to substantially reduce average order delays.