This paper presents a multi-objective optimization method based on NSGA Ⅱ to solve a kind of flexible job-shop scheduling problem(FJSP) under mixed work calendars. A time reckoning mechanism based on the machine's work calendar is proposed and related time reckoning functions are designed. A two-segment encoding method is used to encode the processes and equipments. A two-segment crossover and mutation operator is respectively used to implement crossover operation and mutation operation, in which an improved strategy of genetic operators is applied to ensure feasibility of the progeny individuals and reduce amout of calculation. The time reckoning functions proposed above are used to calculate start time and end time of each operation accurately so as to ensure feasibility of each scheduling scheme in the decoding operation. The following two techniques are employed to shorten production cycle so as to improve quality of each scheduling scheme in the decoding operation:1)Operation time is subdivided into adjusting time and processing time so that the machine of the next operation can be adjusted in advance. 2)A forward extrusion scheduling method is used to arrange each operation so as to reduce idle time of the machines. The research result shows that the proposed method can provide an effective Pareto set of the flexible job-shop scheduling problem under mixed work calendars for the dispatcher.