AN OPTIMAL STRUCTURED POLICY FOR MAINTENANCE OF PARTIALLY OBSERVABLE AIRCRAFT ENGINE COMPONENTS

This paper considers the maintenance of aircraft engine components tha t are subject to stress. We model the deterioration process by means o f the cumulative jump process representation of crack growth. However, because in many cases cracks are not easily observable, maintenance d ecisions must be made on the basis of other information. We incorporat e stress information collected via sensors into the scheduling decisio n process by means of a partially observable Markov decision process m odel. Using this model, we demonstrate the optimality of structured ma intenance policies, which support practical maintenance schedules. (C) 1998 John Wiley & Sons, Inc.