针对综合性面试中的面试官分组决策问题,给出了面试官的分组规则,以此为基础,构建了一个多目标非线性0-1整数规划模型,其目标分别为极小化各组中来自于同一专业(业务或职能)的面试官数量和极小化各组中面试官数量与每组平均面试官数量之间的差异,并开发了一个针对此类问题求解的遗传算法;然后,将面试官分组问题转换为最大流问题,并开发了一个基于Ford-Fulkerson标号法的求解算法。最后,通过一个算例说明两种方法的可行性与实用性,并比较分析了两种方法的优缺点,指出了每种方法的适用范围。
With regard to the grouping problem in comprehensive interviews, this paper firstly proposed the rules for interviewer grouping. Then, based on those rules, a multi-objective nonlinear integer model was built, where one objective was to minimize the number of interviewer who has the same specialty, and another was to minimize the differences between interviewer number in each group and the average interviewer number. To solve the model, a genetic algorithm was developed. In the following, the interviewer grouping problem was transferred into a maximum flow problem, and the solution algorithm based on the Ford-Fulkerson was proposed to solve the maximum flow problem. Finally, the feasibilities and availabilities of the proposed two algorithms were analyzed by a numerical example, from which the application situations of these two algorithms were derived.