多目标粒子群算法在交叉培训规划中的应用

Multi-objective particle swarm optimization algorithm for cross-training programming

为了进一步提高人力资源交叉培训规划的实用性,增加了对于员工学习行为的考虑,提出了在保证任务覆盖水平的基础上,获得员工满意度最大和学习效率最高的多目标优化模型.本文针对问题的特征,采用多目标粒子群(MOPSO)算法对多目标优化模型进行了求解,并设计了多种算法策略,以适应不同的问题环境.通过数值实验,分析了不同问题规模下,针对不同性能指标算法参数和策略的适用性.最后,以柔性单元装配生产线为例,进行了数值实验,实验结果表明了模型的有效性和合理性.

In order to improve the practicability of a cross-training programming, the factor of human learning behavior is considered. A multi-objective optimization model is presented on the basis of task redundancy policy, in which the objective functions describe the labor satisfaction and the learning efficiency. A cross-training programming based on multi-objective particle swarm optimization algorithm （MOPSO） is proposed. The MOPSO solves for the solutions of the proposed multi-objective optimization model and designs algorithm policies for different problem environments. Several flexible cell assemblies in different scales are presented for modeling the environment in a series of numerical experiments. Results in each environment are analyzed in the ,~spects of diversity, distribution and convergence index. The analyzed results show that the method presented in this paper can solve cross-training programming problems effectively.