The main task of output module is to output the schedule to the external file with the human-readable format. In scheduler module, we use a integer to represent a course, a candidate and a professor. Now we should map this integer to the string name which can be viewed by users. This map information is provided by the input module. Currently output module will produce 3 text files for each schedule. One is the candidate file, each line includes such fileds: candidate name, two coures, three professors' name, exam date, exam time and exam location. Each line is corresponding to one session, so each candidate exactly occupies three lines. The other file is the faculty file, each line contains the professor name, exam date, exam time, exam location and two courses in this session. The number of lines one professor appears means the total sessions he need to be attended. The professor just look at this file to determine when and where to attend a session. The third file is a workload file which contains each schedule's professor workloads and some basic information of the schedule. Each line of which is a professor name and his workloads. At the end of this file there are some basic information of the schedule including the minimum timeslots, the minimum competence of sessions and the average competence of sessions. Since our scheduling algorithm is sort of random, basically we will run thousands of times and export some schedules. Users will choose a "best" schedule by hand based on the information of workload file.