基于案例决策理论的仿真任务共同体服务选择算法

资源分配优化是面向服务（ SOA）的大型多用途仿真系统的关键问题，而含有不确定状态或效果的服务选择是资源分配优化中的一个难点。针对以上问题，提出了构建仿真任务共同体的方法，基于案例的决策理论（ CBDT）的思想，设计了一种不确定型仿真任务共同体服务选择算法（ CBDSSA）。算法通过服务选择案例相似度的层次化运算，生成相似历史案例集，计算得到相似历史案例的方案效用值，最终评估出目标服务选择方案综合效用值并排序备选。算例验证表明，算法约束条件少，层次结构分明，运算结果直观，为仿真任务共同体服务选择问题提供了一种新的思路和实践手段，对大型多用途仿真系统资源分配优化研究具有一定的参考价值。

Study of Simulation Task Community Service Selection Algorithm Based on Case-based Decision Theory

Resources allocation and optimization becomes a key problems during the research of service?oriented large?scale multipurpose simulation system, in the meanwhile service selection with unknown status is one of the difficulties in resources allocation and optimization. To solve the problems, the paper proposes a method of simulation task community, designed a task community service selection algorithm ( CBDSSA) based on case?based decision theory. By the means of coherent calcu?lation, the algorithm can successfully build similar historical case set and create the utility of its implementation effect. Final?ly, the overall utilities of each alternative service selection cases are calculated and sorted by integrating the similarities and utilities. The algorithm needs fewer constraints and has strong logicality and practical applicability. It provides a new method of practice in task community service selection, and has reference value for the research of service selection of large?scale multipurpose simulation system.