Functional importance evaluation approach for cloud manufacturing services based on complex network and evidential reasoning rule

Abstract

In the uncertain cloud environment, manufacturing services monitoring is effective to ensure the normal function of cloud manufacturing service system (CMSS). However, continuously monitoring all of the manufacturing services is impractical since it is resource consuming. One feasible method is to prioritize the allocation of monitoring resources to the important services. Therefore, we propose an approach for evaluating the functional importance of manufacturing services based on complex network and evidential reasoning (ER) rule. Firstly, a domain-oriented cloud manufacturing service complex network (DoCMSCN) model is constructed and elaborated. Secondly, based on the idea of multi-granularity and multi-indicator, an evaluation model for the importance of manufacturing services is presented. Different centrality indicators of the DoCMSCN in different functional granularities are obtained and transformed into evaluation evidence. Then the ER algorithm is applied to fused the evaluation results. In the fusion process, the reliability and weight of each piece of evidence are fully considered to improve the fusion accuracy. The experimental results of vertical elevator design services show the proposed approach can effectively find the important manufacturing services and superior than the existing ones. It can facilitate the decision-making of monitoring strategy in conditions of limited resources from the functional perspective. Finally, we develop a prototype monitoring system for vertical elevator design services.