Order sequencing for a bucket brigade seru in a mass customization environment

In a mass customization environment, customer demands may change dramatically, which challenges conventional production systems. Seru production system (SPS) is a key to deal with changing varieties and fluctuated volumes. A bucket brigade seru is a divisional seru, in which partially cross-trained workers are organized as a bucket brigade. This paper studies an order sequencing problem for a bucket brigade seru with the objective of minimizing makespan. Dynamic behavior of a bucket brigade seru is modeled. Properties of the problem are presented that alternating the processing sequence of some products does not affect makespan. These properties provide factory managers with flexibility to adjust products processing sequence in case of changes in customer demands or lacks in materials. Six local search algorithms are developed using these properties, which combine three policies of selecting products in the operations of swap and insert. A bucket brigade local-search-based differential evolution (BLaDE) algorithm is proposed for the addressed problem, which may incorporate one of the six local search algorithms according to various scenarios. More than 14,000 instances are generated to evaluate the algorithms in terms of effectiveness, convergence, and stability. Experimental results demonstrate the superiority of BLaDE in volatile markets.