Abstract
Shoes are the human life necessities and the consumption is about 12 billion worldwide every year. Along with gradually increasing population all over the world, the potential footwear market is substantial. Due to the variable types of products, there are difficulties in frequently redesigning the production line. Furthermore, stitching line which is highly dependent on operators, needs numerous manual operations. Thus, by given the fixed number of operators and equipment, deciding the optimal assignment to workstations becomes more important. This research proposes a Non-dominated Sorting Genetic Algorithm-II (NSGA-II) to solve multi-objective assembly line balancing problem. Two criteria are simultaneously considered for optimization: to minimize the cycle time, and to maximize the workload smoothing (i.e. to distribute the workload evenly as possible to the workstations of the assembly line). It compares between simple and complex models with different number of tasks, in which the data is collected from a footwear manufacturing company in China. In this study, design of experiments (DOE) is adopted to validate the performance and analysis of variance (ANOVA) to evaluate the output. Moreover, the constrained labors and equipment, the proposed approach can provide a practicable model in real application and increase the production efficiency.