系统工程与电子技术 ›› 2018, Vol. 40 ›› Issue (5): 1057-1063.doi: 10.3969/j.issn.1001-506X.2018.05.15

• 系统工程 • 上一篇    下一篇

基于rollout策略下的决策实体配置问题求解方法

廖梦琛1, 孙鹏1,2, 张杰勇1, 武君胜3   

  1. 1. 空军工程大学信息与导航学院, 陕西 西安 710077; 2. 西北工业大学计算机学院, 陕西 西安 710077;3. 西北工业大学软件与微电子学院, 陕西 西安 710077
  • 出版日期:2018-04-28 发布日期:2018-04-24

Solving method for decision-makers configuration problems based on rollout strategy

LIAO Mengchen1, SUN Peng1,2, ZHANG Jieyong1, WU Junsheng3   

  1. 1. Information and Navigation College, Air Force Engineering University, Xi’an 710077, China; 2. School of Computer Science, Northwestern Polytechnical University, Xi’an 710077, China; 3. School of Software and Microelectronics, Northwestern Polytechnical University, Xi’an 710077, China
  • Online:2018-04-28 Published:2018-04-24

摘要:

针对传统层次聚类法采用贪婪策略的聚类过程可能无法达到聚类效果最优的情况,提出了一种基于rollout策略下的层次聚类法对所得聚类方案进行优化改进。分析了聚类过程中决策实体、平台与任务之间的关系以及约束条件,以作战任务的执行时间作为工作负载测度,建立以决策实体工作负载的均方根(root mean square, RMS)为目标函数的问题数学模型,以任务与平台的分配关系作为输入信息,在基于最小RMS值的平台合并准则下采用rollout策略对层次聚类法的每层聚类进行优化,得到平台与决策实体的优化配置关系。最后通过联合作战仿真算例和一般算例进行仿真分析,验证了该方法的可行性和优越性。

Abstract:

To solve the problem of decision-makers (DM) configuration led by traditional hierarchical clustering algorithms under the greedy strategy, an improved hierarchical clustering algorithm based on the rollout strategy is presented. On the basis of analyzing the relationship and constraints among the DM entity, the platform and the task, the processing time of the operational task is used to measure DM’s workload, and the mathematical model of root mean square (RMS) of DM’s workload as the objective function is established. Relationship between the task and the platform is input, the rollout strategy is used to optimize the hierarchical clustering of each layer under minimum RMS merger criterion, and the optimal configuration relationship between the platform and the DM is obtained. Finally, the feasibility and superiority of the algorithm are verified by a case of joint campaign and a general case.