A framework for MGrid resource service optimal-selection and composition

The problem of resource service optimal-selection and composition (RSOSC) in manufacturing grid (MGrid) system is proposed. Based on the authors' previous work, a framework for addressing RSOSC problem in MGrid system is presented. The framework consists of five layers: (1) task layer-responsible for MGrid task decomposition, (2) resource service match and search layer-responsible for searching the qualified resource service according to each decomposed subtask's requirements, (3) resource service QoS processing layer-responsible for collecting QoS information for each candidate resource service, as well as QoS evaluation, comparison, monitor, updating, (4) resource service optimal-selection layer-responsible for evaluating and ranking the candidate resource service for each subtask, and (5) resource service composition layer- responsible for resource service composition and optimal-selection.