The blackboard model of problem solving
AI Magazine
Reasoning with worlds and truth maintenance in a knowledge-based programming environment
Communications of the ACM
| Hi-index | 0.00 |
A number of problems such as planning, scheduling and system maintenance have in common a dynamic set of resources and the need to optimize the use of these resources. The Dynamic Optimal Resource Utilization System (DORUS) is an architecture suitable for the construction of expert systems for these problem domains. It allows not only for dynamic data but for problem solving strategies which might change according to time or circumstance. A DORUS-based system will respond to a predefined set of events. Events may take the form of some change to the set of resources or the passage of time.When a response to an event is required, an appropriate goal is selected and resources are allocated according to a set of rules and heuristics. The set of rules which is activated is determined by the goal and current circumstances. The system is capable of generating a plan to implement the suggested allocation of resources. Finally, a DORUS-based system allows “what if” scenarios so that the consequences of unexpected events or alternative strategies can be explored.In this article we describe the DORUS architecture and present as an example an expert system for the reconfiguration management of flight data. Other potential applications are also discussed.