A collaboration mechanism for service-oriented manufacturing processes with uncertain duration: a perspective of efficiency

Abstract

ABSTRACT The Industrial Internet is widely leveraged to connect diverse manufacturing resources, thus manufacturing allies can be dynamically organized to meet market opportunities. Manufacturing resources at various levels are virtualized in the form of manufacturing services and the core manufacturers can organize their manufacturing processes by choosing appropriate services to complete certain manufacturing tasks. When multiple services are orchestrated together, a collaboration mechanism is crucially important for the performance of the service-oriented manufacturing processes. However, many uncertain factors punctuate services from performing in a sound rhythm. In this paper, the authors analyze the factors that have impacts on the service to complete a certain task, establish a framework of the mechanism for service collaboration, propose an efficiency measurement method for the collaborative service processes, and construct a service collaboration model considering service duration, task relation and set probability. A Monte Carlo simulation improved Particle Swarm Optimization algorithm is developed to solve the model. An experimental study has been made with a car manufacturing process to verify the mechanism and the results show that the overall performance will be improved when services collaborate with this mechanism.