Design of a hybrid cloud manufacturing environment for production of personalized products

Abstract

The growing demand for highly customised products has seen manufacturing companies seek new ways to streamline their internal business processes, reconstruct their production lines, and reengineer their business models to increase responsiveness in a dynamic market. As such, the effective management and utilisation of global manufacturing resources are crucial to achieving customer satisfaction and enabling companies to deliver fast time-to-market customised products at changing production scales. However, manufacturing companies, especially Small and Medium-sized Enterprises (SMEs) have limited resources and experience, and lack effective systems and tools to work in an ever-increasingly distributed, networked cloud environment that calls for significant production flexibility. This results in a need for better resource management and production planning in a global production network. In response to this need, the MCloud project was begun. The key objective of this project was to develop an integrated solution to enable rapid production of highly customised products, using globally sourced manufacturing resources. A flexible, web-based system (MCloud) was developed with the aim of enabling rapid configuration of distributed manufacturing resources through better understanding of resource constraints for the production of highly customised products. The overall system model is inspired by an emerging service-oriented manufacturing paradigm – cloud manufacturing. Cloud manufacturing is a serviceoriented, knowledge-based new mode of networked manufacturing. It aims at facilitating manufacturers to minimise their product life-cycle expenses, maximise production efficiency, and provide agile accommodation of the available manufacturing assets to the variable demands of customers. Cloud manufacturing assumes that manufacturers use crowdsourcing and outsourcing models for manufacturing operations and supporting services. To determine the significant research dimensions to be considered, research was conducted on cloud manufacturing, and collaboration was established with several industry partners. The system developed addresses the research gap of implementation of theoretical cloud manufacturing research in practical applications. This is achieved by providing industry users with a hybrid cloud manufacturing environment to facilitate effective resource sharing and rapid product development. The final MCloud framework consists of four major system modules that collaboratively enable highly customised manufacturing jobs to be quickly processed by distributed manufacturing resources in the cloud. The implementation of the proposed system provides manufacturing companies of all sizes with an effective solution to managing manufacturing resources and processes in a highly customisable environment. The performance of MCloud system has been validated through case studies with industrial partners. Valuable capabilities of the system include: (i) service-oriented representation of highly customised products using a standardised product data model; (ii) effective capability modelling of manufacturing resources using a multidomain, multi-level and multi-granularity resource virtualisation mechanism; (iii) explicit specification of context-aware access control policies for any cloud resources, and establishment of a dynamic hybrid cloud manufacturing environment; and (iv) rapid production planning for a customised manufacturing job using knowledge-based service composition and availability-adaptive plan adjustment. The research work in this thesis has been reported in four journal papers and presented at four international conferences. Keywords: Cloud manufacturing, Mass customisation, Service-oriented manufacturing, Semantic web, Expert system, Hybrid cloud