川端 亮

Single and parallel structure servers, although powerful mathematical models in queueing theory, do not address a wide range of real-life systems. In this paper we propose a composite-server model and make use of the knowledge of the intrinsic composition of its service providing units (personnel or equipment) to arrive at Qualitative Rules needed to evaluate its performance. The composite server model has wider scope in its applications and can be used to represent a variety of system classes. We use this novel concept in the performance design and improvement of collaborative engineering systems. A frequent source of bottleneck in collaborative systems is the lack or surplus of service-providing units, known as "Perspectives" in the Multi-Context Map (MCM) modelling terminology. Bottlenecks due to inappropriate Perspective allocation are resolved by the Qualitative Reasoning approach.

Although detailed flows are not always well-documented in business practice, the relevant activities of organizations and stakeholders, as well as associated state transitions caused by them, can easily be captured in use case analysis. We have developed a method of designing a conceptual data mode by analyzing Petri nets of the system from the viewpoint of actors in use case diagrams. The main feature of our method is that a conceptual data model can be constructed even if we do not collect detailed data items of the business process. Moreover we can find necessary data in the business flow even if these data are not present in the data item list that was initially collected.

To make collaboration tasks work effectively, it is very important to analyze what collaborators are doing, which position they are in, and how they communicate with each other. Multi-Context Maps (MCM) and Collaborative Linkage Maps (CLM) together constitute a solution for such demands. These methods analyze each collaborator's position precisely, so these models capture important information in the process. In this paper, we propose to extract E-R models, which are widely used to design databases in the early phases, from MCM and CLM. The process of composing E-R models still relies on the designers' experience or their intuition, which often causes inadequate formalization of complex requirements. Our extraction provides a systematic method of designing data models. We also propose how to apply the extracted model when designing databases.