Assignment and configuration of multi-skilled worker is a key problem for Seru production systems formation, which is a human-centered production system. Considering the selectivity of worker skill sets and the diversity of workers' proficiency level, this paper develops a multi-objective optimization model of the multi-skilled worker configuration problem for the Seru production system to minimize the make span and imbalance in labor hours. The algorithm based on NSGA-II is developed to solve the mathematical model considering the NP-hard and multi-objective optimization nature of the problem. At the same time, numerical experiments are conducted to evaluate the impact of varying upper bound task number and proficiency level on make span, workload imbalance, average skill number and average Seru number.