Dynamic scheduling of manufacturing processes using dimensional databases

Abstract

On-line scheduling has traditionally been restricted to single machines or work cells. In this paper, a dynamic on-line scheduling algorithm is presented to encompass several machines or work cells. The algorithm uses a dimensional database to store and query system information such as machine availabilities and other requirements. Due to the built-in capabilities of the dimensional data model for the seamless handling of dimensional data, the algorithm does not depend on a particular data type. As illustrated in this paper, this is particularly applicable to flexible manufacturing systems, where the system data may include any assortment of dimensional data. Such data may include ordinary/temporal data about resource availabilities, spatial data about machine locations and spatio-temporal data such as prior schedules.