基于神经网络辅助的智能人员排班系统  

Intelligent Personnel Scheduling System Based on Neural Network Assistance

作　　者：陈子夷 豆亚杰 姜江 杨克巍[1] 谭跃进[1] CHEN Ziyi;DOU Yajie;JIANG Jiang;YANG Kewei;TAN Yuejin(College of System Engineering,National University of Defense Technology,Changsha Hunan 410072,China)

机构地区：[1]国防科技大学系统工程学院,湖南长沙410072

出　　处：《指挥与控制学报》2023年第2期215-224,共10页Journal of Command and Control

基　　金：国家自然科学基金(71901214);国防科技大学“高层次创新人才培养计划”资助。

摘　　要：为快速获得优质的排班表,设计了结合深度神经网络和分支定界法的智能人员排班系统.介绍人员排班问题的特点和难点;构建问题的整数规划模型;提出基于深度神经网络辅助的分支定界法,通过学习现有的已知最优解的人员排班问题,在分支定界的每一步作出合理的分支选择和修剪.该方法是使用深度学习方法解决组合优化问题的一种创新尝试,实验部分针对不同的参数设置和标准实例集合验证了该方法的可行性.In order to obtain a high-quality schedule,an intelligent personnel scheduling system combining deep neural network and branch and bound method is designed.The characteristics and difficulties of personnel scheduling problem are introduced;the integer programming model of the problem is constructed;a deep neural network-assisted branch and bound method is proposed.It can make reasonable branch selection and branch pruning at each step in branch and bound by learning the existing personnel scheduling problem with known optimal solutions.The proposed method is an innovative attempt to solve combinatorial optimization problems with deep learning method.The experimental part verifies the feasibility of this method on different parameter settings and standard instance sets.

关 键 词：人员排班 组合优化 深度神经网络 分支定界法 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]