Preparation and support of scheduling employing knowledge acquisition, planning, and rule-based decision making

Abstract

Over the last 30 years, much work has been done to develop methodologies that solve scheduling problems. What is missing most often in these systems is user support for preparing plan generation and formulating the knowledge that describes a scheduling problem. The authors propose a knowledge-based support paradigm in which the following domains from expert systems technology are transferred to the field of scheduling: knowledge acquisition methods, planning techniques, and rule-based decision support. We have developed a component for supporting the acquisition of knowledge relevant for scheduling. This component collects reusable information and checks consistency of added knowledge with existing information. Another component uses a knowledge base to support the user, for example, a project manager, in composing the set of actions and resources that are required for a task. The planning process is not goal-driven, but rather an interactive construction process. Finally, a decision support approach is presented for relaxing constraints of actions competing for limited resources. The knowledge base is equipped with a comprehensive set of conflict resolution alternatives and a set of rules that can guide the resolution process.