Expert Database Systems: Efficient Support for Engineering Environments.

Abstract

Manufacturing and Engineering processes use both large scale data and knowledge bases, and the use of expert systems in such environments has become a necessity. Expert Database Systems have evolved from conventional database systems to meet the requirements of current Artificial Intelligence applications. However, future Expert Database Systems will contain knowledge bases of significant size which makes main memory insufficient and the use of a database system a necessity. We propose an effective way of building High Performance Expert Database Systems to support manufacturing and engineering environments. These systems are based on Incremental Computation Models; such models utilize results of previous computations by merging them with newly derived results of computations on small increments representing changes in the database. Our system will be able to support very large knowledge bases by utilizing novel structures and acceas methods and by using a very sophisticated inference engine based on incremental computation models.