林家柔

Virtual research and development (R&D) is inevitable to reduce the product life cycle. Enterprises tend to rely on their foreign partners for supporting technology and knowledge acquisition to conduct and improve firms’ product development with low R&D risk. R&D is a highly creative and knowledge-intensive activity, Therefore, efficient knowledge flow, which transmits the right knowledge to the right people at the right time, is key to improving efficiency of the R&D process.

kanban Management has been recognized as an important tool for the implementing lean production. Kanban supports visual production control using the card of providing information to regulate the flow of inventory and materials. To enhance the knowledge flow efficiency in the virtual R&D process, this study proposes a knowledge kanban system utilizing the philosophy of kanban management and knowledge engineering techniques. The knowledge kanban system provides visual signals to employees about relevant knowledge that is necessary to support them in executing their activities smoothly and the knowledge they need to create, share, and revise.

To achieve this objective, this study first proposes a virtual R&D model and a knowledge flow model in virtual R&D based on the analysis result of knowledge in VEs and the VE life cycle, and then designs the knowledge kanban operational model and the knowledge kanban representation model according to the knowledge flow model in virtual R&D. Finally, this study proposes the knowledge kanban functional framework and develops the related technologies to implement the knowledge kanban system. The knowledge kanban system is an effective tool to facilitate knowledge creation, storage, transmission and sharing for R&D engineers to develop knowledge in problem solving and product development, to improve enterprise competitiveness.

Global competition and the rapid development of information technology make virtual enterprise one of the most promising business strategies. The success of virtual enterprising heavily depends on the integration of business workflows across enterprises. However, besides the problems of differing data formats, time variation, and system heterogeneity, the multi-enterprise workflow integration faces difficulties due to the traditional workflow design’s lacking of flexibility and dynamicability. Therefore, how to plan workflow for virtual enterprise rapidly and correctly to shorten the lead time is the motivation of the research. 　The purpose of this research is to develop an automatic workflow planning model and mechanism for virtual enterprising. The model includes two phases, i.e. process planning and process verification. In the phase of process planning, HTN planning algorithm is first utilized to find out activities and tasks that meet the process goal. A proposed sequencing algorithm is then developed to connect related activities and tasks into a complete process. In the second phase, a verification algorithm is designed to verify the workflow planned in the previous stage to avoid mistakes. 　The proposed automatic workflow planning mechanism can be applied to the B2B applications. It separates the workflow system from the enterprises to let the workflow system work independently, that makes the planned workflow more adaptable and reduces the number of controlling logics, and thus releases the burden of workflow management. The result of this research can shorten the time to plan the virtual enterprise’s workflow, increase the flexibility and efficiency of the workflow system, and reduce the exception handling cost, and consequently increase business performance of virtual enterprises.