DECISION-THEORY IN MAINTENANCE STRATEGY FOR A 2-UNIT REDUNDANT STANDBY SYSTEM

The purpose of this study is to obtain a good maintenance strategy for a standby system. The problem is analyzed via decision theory to dete rmine the waiting time to call the repair facility (for a 2-unit stand by system) when the first piece of equipment fails. Previous research into this kind of system is briefly described, and a need for construc ting a decision model is explained. The uncertainty of the parameters are accounted for in a Bayes approach in order to consider expert prio r knowledge. The failure & repair rates are assumed to be constant and are elicited from an expert's prior beliefs. When no data are availab le, then expert guesses are used. This paper presents a method for sol ving the conflicting requirements of system availability and cost thro ugh a multi-attribute utility function which can express cardinal valu es for the decision maker's preferences over the objective variables. The decision model derives the appropriate maintenance strategy; it co rresponds to a set of actions, procedures, and resources, giving a con sequent waiting time before calling the repair facility. The use of th e model is demonstrated for a telecommunication system, including a se nsitivity analysis related to prior probabilities of failure & repair rates; the robustness of the procedure is verified.