Abstract: As the amount of manufacturing resources in cloud manufacturing environment is huge and widely distributed, for business process, reasonable resources composition can improve the efficiency of resource selection and utilization ratio. A correlative matrix according to the dependencies between different manufacturing resources is built, which is allocated to business activities in workflow model. It uses the algorithm of bond energy to do binary partitioning to the correlation matrix, and composes the resources which have stronger dependency to each other, so as to form a number of resource sets. Taking product design, manufacture and assembly as example, an experiment is done and a performance comparison for resource selection shows that the proposed method is available.

Key words: cloud manufacturing, resources composition, correlation degree, workflow

摘要： 云制造环境中资源数量巨大，且广泛分布，对于业务过程而言，合理的资源组合可提高资源的选取效率以及利用率。以工作流模型为基础，根据资源之间的依赖关系构造关联度矩阵，进而采用键能算法对关联矩阵做二分变换，将依赖关系强的资源聚集到一起，形成若干个资源集。以产品设计、加工和组装为例，通过实验以及对资源选取效率的对比，验证该方法的有效性。

关键词: 云制造, 资源组合, 关联度, 工作流