Modelling and a hybrid genetic algorithm for the equity-oriented worker assignment problem in <i>seru</i> production systems

Worker assignment is critical for developing a flexible and human-centered seru production system (SPS). Most studies in the literature focus on worker assignment with the objective of maximizing system efficiency, which causes an inescapable consequence that workers with high proficiency levels experience overwork compared with workers with lower proficiency levels. From the perspective of workload equity, we investigate the worker assignment problem in SPS, where the heterogeneity of workers with diverse skill sets and skill proficiency levels is taken into account. A mixed integer nonlinear programming model named the equity-oriented model is proposed for this problem to mitigate the workload inequity among workers by minimizing the maximum workload of workers. The equity-oriented model is linearized and solved by the CPLEX solver for small-scale instances. For efficiently solving large-scale instances, we design a hybrid genetic algorithm combined with local search. Computational experiments demonstrate the high performance of the hybrid genetic algorithm, both in terms of computing time and solution quality. We find that using the equity-oriented model can significantly balance the workload among workers at the cost of minimal increases in the total labor hours.