Abstract: To strengthen regional collaborative manufacturing and to realize service-oriented manufacturing,an adaptive organization and finding mechanism of Multi-granularity Manufacturing Resource (MMR) based on geographic region and Manufacturing Resource Classify Tree (MRCT) was proposed.The MMR model and the multi-granularity manufacturing service model based on business process were constructed to be the basis of collaborative resource storage,and then the organizational structure of MRCT and MMRs were discussed.Furthermore,the adaptive mechanism of MMRs'organizational structure was discussed,including dynamic joining and exiting of MMR and MMRs'maintenance,splitting and merging of resource node based on load sensing,and MMR's discovery algorithm based on the retrieval and the three phase matching was designed.Through the comparison of algorithms performance,the classification retrieval algorithm based on MRCT had faster searching performance than similar algorithms,and multi-granularity service had faster than combined service by several single granularity services in resource discovery performance.The validity of proposed model and related algorithms were verified through the simulation experiment.

Key words: multi-granularity manufacturing resource, region collaborative manufacturing, manufacturing resource classify tree, adaptive organization structure, discovery algorithm

摘要： 为了加强区域协同制造,实现面向服务的制造,提出一种基于地理分区与制造资源分类树的多粒度制造资源自适应组织与发现机制。构建了基于业务流程的多粒度制造资源和多粒度制造服务模型作为协同资源存储的基础,讨论了制造资源分类树和多粒度制造资源的组织结构,进而给出了多粒度制造资源组织的自适应机制,包括多粒度制造资源的动态加入和退出以及资源的维护、负载感知的资源节点的分裂与合并,并设计了基于分类检索与三阶段匹配的多粒度制造资源发现算法。经算法性能比较,基于制造资源分类树的分类检索算法比同类算法有较快的搜索性能,粗粒度服务比单粒度服务组合具有更快的资源发现性能。通过仿真实验,验证了模型及其相关算法的有效性。

关键词: 多粒度制造资源, 区域协同制造, 制造资源分类树, 自适应组织, 发现算法