The bus crew scheduling problem (BCSP) is to assign drivers and conductors to daily crew duties that cover a set of schedules for buses of a public bus company during a period of time, e.g., a week. By combining case-based reasoning (CBR) and rule-based reasoning (RBR), this paper introduces the design and implementation of a bus crew scheduling system (BCSS). The current system works like this. Firstly, an inference mechanism integrating CBR and RBR is designed to obtain near-optimal solutions. Secondly, based on these near-optimal solutions, a genetic algorithm is developed to produce better solutions. The practical applications show that the system is effective.