A distributed problem-solving approach for owner–contractor prequalification

This paper reports the results of work that aims to study the practical implementation of Distributed Artificial Intelligence (DAI) and to introduce its capabilities in representing and using knowledge in the area of contractor prequalification. This is accomplished by developing an intelligent Knowledge-Based Decision Support System for a contractor prequalification environment. The resulting DAI architecture comprises a hierarchy of loosely coupled problem solvers, all operating under the supervision of a top-level control mechanism. Each problem solver works as a classifier system. The problem-solving knowledge for each problem solver is developed using (1) machine learning by training neural networks, and (2) heuristic rules of thumb. A learning and refining subsystem is attached to the system for refining the existing knowledge to improve the system performance.