In medical outpatient services, due to patients’ imbalanced selection for doctors of different levels and for different visiting periods, inefficiency of resource utilization and dissatisfaction of patients have become the main problems faced by hospital managers. For the first time, this research has considered patients’ preference between high-ranking professional titles of general doctor and expert doctor. Through analyzing real data of the outpatient clinic at Dalian City Dermatology Hospital, the behavioral pattern of patients’ patience limit adjusted with expected waiting time was obtained. This research also established a data-driven discrete event simulation model that takes into account walk-in patients’ time preferences, appointment patients’ no-shows and cancellations, and considers complex patient flow caused by unbalanced selection of doctor resources and patience limit of waiting time. To optimize scheduling for appointment patients with two types of doctors, this research put forward a simulation optimization framework that maximized hospital benefit and minimized patients’ dissatisfaction. At the same time, simulation budget allocation based on multi-objective optimization and genetic algorithm were combined to obtain the approximate Pareto joint capacity plan of multi-servers and a patient scheduling scheme. The simulation model was validated through a case study based on real data of outpatient service for the whole year, and the proposed optimization method can comprehensively improve performance of outpatient service scheduling system. The simulation optimization framework can provide an effective scheduling scheme for all multi-server service systems involving consumer selection and impatient behavior.