INTEGRATED DIAGRAMMATIC REPRESENTATIONS FOR DATA DESIGN IN COLLABORATIVE PROCESSES

摘要

To make business processes work effectively, it is critical for workers to communicate and collaborate with each other. Especially, when designing supporting systems for collaborative, processes, the business experts and the professional systems designer have to communicate effectively, though they have different viewpoints to the system. Business experts prefer capturing general view of the entire system. On the other hand, system designers require detailed information without ambiguity. This fact sometimes results in communication problems. To solve this problem we propose to integrate the methods developed for each of them to facilitate the design process. We introduce two methods, IDEFO and Multi-Context Map (MCM). Both methods are designed to analyze collaborative processes and produce function models, though the target user is quite different. IDEFO is a simple and a flexible method, and is mainly used as a communication tools among collaborators who are not familiar with complex notations, such as business experts. On the other hand, MCM produces a precise description, and is for professional system designers and engineers, who require detailed information without ambiguity. MCM has richer syntax and defines the collaboration tasks in detail. MCM is powerful when formalizing complex requirements, though some collaborators may prefer simple methods they are used to, such as IDEFO, to MCM. To support the exchange of information between MCM and IDEFO, we develop a bi-directional converter between them. In this paper, we propose two methods of designing E-R models that is useful for database design between business experts and professional system engineers. In many cases, the process of designing databases still relies on professional system engineer's intuition. This fact sometimes causes the information among business experts not to be captured properly in the requirements because of the communication problems between business experts and professional systems engineers. Our methods facilitate the process of sharing information between them, and enable the designer to reflect the information properly in the requirements. One of the methods is extracting E-R models from well-disciplined IDEFO diagrams to facilitate business experts in designing databases, and the other is from both MCM and CLM, which enable the systems designer to capture detailed data structure of the databases. The function models enable the designers to define the scope of the system effectively, which is very difficult by using only the data models. We conclude our paper by evaluating our methods.