Adaptiveness in information systems integration
Gilbert Babin · 1993
Interoperability, local autonomy, and concurrent processing are major problems facing the integration of multiple data and knowledge systems, such as those found in today's manufacturing enterprises. Their solution, however, requires new analyses and insights beyond the previous understanding and formulation of these issues. We submit that the key to solving these problems is adaptiveness. The objective of this research, therefore, is to develop the new capabilities of adaptiveness needed to enable the enterprise to grow (i.e., incorporating new or legacy systems into the environment) and change its integrated information system while avoiding the well-known performance problems caused by the traditional approaches to global control and management. The solution utilizes (1) a metadatabase serving as the knowledge base for enterprise integration, (2) a concurrent architecture using the metadatabase for integration, and (3) a rule-oriented programming environment (ROPE) for implementing and managing the concurrent architecture. The technical nature of the research problem is concerned with concurrent knowledge representation and processing. This knowledge method must be able to (1) control the application logic of local systems without the use of a central controller, (2) enhance the local applications with a distributed knowledge capability for global behavior, and (3) transfer the needed knowledge both between the metadatabase and the local applications, and among the local applications themselves. The new ROPE method contributes not only to the metadatabase solutions to the problem of adaptiveness, making the metadatabase an active heterogeneous distributed database management systems, but also to the general concepts of: local autonomy, open system architecture, interoperability, and systems evolution. Together, the research extends the current information integration technology and results in a particular method effecting adaptiveness in managing multiple systems environments. Moreover, its specific generic results provide the architecture, algorithms, and languages to further the state-of-the-art of current knowledge representation and processing, as well as the shells design which affords an efficient model for application program interface systems in general.