Artificial intelligence and decision analysis

Much work has been done on the development of modelling techniques to assist in the process of decision-making. The areas of operational research and statistical analysis are well established, and the speed and memory capacity of computers has facilitated the solution of very large problems using such techniques. Recent developments in artificial intelligence, particularly with regard to expert systems, open up the possibility of modelling the decision-maker rather than the decision situation. This would result in a much more sophisticated decision-support system, but as yet it is an approach which has not been investigated to any significant degree. This paper suggests how AI techniques might be combined with traditional problem-based methodologies to provide a comprehensive decision-analysis support-system framework, modelling the activities of a human decision-maker and providing access to a host of traditional and artificial intelligence tools which can be effectively interfaced.