The case-library method for service composition and optimal selection of big manufacturing data in cloud manufacturing system
The case-library method for service composition and optimal selection of big manufacturing data in cloud manufacturing system
复制标题
DOI：
10.1007/s00170-015-7813-8
复制
发表时间：
2015-09
期刊：
影响因子：
--
通讯作者：
Feng Xiang;Guozhang Jiang;Lulu Xu;NianXian Wang
中科院分区：
文献类型：
--
作者：
Feng Xiang;Guozhang Jiang;Lulu Xu;NianXian Wang
服务组合与优化选择（SCOS）技术通过组合基本服务来满足不同用户的需求，实现企业资源的有效配置和价值最大化。由于云制造资源池规模的急剧增加，用户需求的日益复杂，使得组合服务的数量、类型、维度和复杂度急剧增加，为云制造的应用和实现带来了大数据环境，本文分析了未来大数据环境下SCOS的难点和解决方案，特别是针对大规模组合服务执行路径（CSEP）的优化选择问题，提出了基于实例库的两阶段SCOS方法。首先从案例库中查找与用户需求相似的案例;其次，利用实例对现有的优化算法进行初始化，以解决大规模的优选问题。研究了案例库结构、用户服务请求结构、相似度比较和实现过程。与现有的优化算法方法相比，在大规模CMFG中使用实例库的原型系统可以获得更好的优化结果。
Service composition and optimal selection (SCOS) technology composes basic services to satisfy various users’ needs and realize enterprise resource efficient allocation and value maximization. Since sharply increasing scale of the cloud manufacturing (CMfg) resource pool, and the growing sophistication of user requests, will make composed service a sharp increase in the quantity, type, dimension and complexity, which cause a big data environment for CMfg’s application and realization, this paper analyzes the difficulties and solutions of SCOS of big data in the future, especially for optimal selection from large-scale composed service execute paths (CSEP), and proposes two phases SCOS method based on case library. Firstly, cases, similar with users’ request, are searched from case library. Secondly, the cases are used to initial the existing optimization algorithm to solve large-scale optimal selection problem. Moreover, case library structure, user’s service request structure, similarity comparison, and realization process are studied. Compared to the existing optimization algorithm method, the prototype system using case library in large-scale CMfg could obtain a better optimization result.