Abstract: As the core resource for enterprises to implement MRO（Maintenance, Repair and Overhaul） services, the scheduling scheme of service workers directly determines the operation efficiency and cost of MRO services. At present, with the personalization and differentiation of MRO customer needs, along with the geographically dispersed nature of demand sites, the previous scheduling theory shows obvious limitations, and it is urgent to carry out targeted theoretical research on scheduling decision-making. Therefore, a scheduling problem for MRO service workers that involves multi-site complex equipment and takes into account multi-class customer constraints is proposed. Firstly, we categorize customers into three distinct categories based on their warranty attributes and recognition levels: in-warranty customers who are still within the warranty period, general customers who possess a certain level of recognition but are beyond the warranty period, and temporary customers who have zero tolerance for delays. A mathematical model is formulated with the dual objectives of maximizing total service revenue and minimizing the maximum completion time simultaneously. Secondly, an enhanced algorithm named MNSGA is introduced, building upon the NSGA-II algorithm, to tackle the aforementioned problem. To optimize the performance of MNSGA, initialization rules and local search operators have been carefully crafted to align with the unique characteristics of the problem. Lastly, based on the test set and the actual data, multiple sets of examples of different sizes are constructed,and a series of experiments are designed. Numerical results show that the proposed operator can effectively improve the performance of MNSGA. Compared with other well-known algorithms, the effectiveness and competitiveness of MNSGA are verified.

Key words: MRO services, customer segmentation, resource scheduling, multi-objective optimization