An Approach to the Minimization of Misclassification in the Repair of Equipment

A problem which can be quite costly to an agency involved with repairable equipment on a large scale is the misclassification which can arise when repair work is initiated. Often, a piece of equipment finds its way to the repair shop even though it has not experienced a failure. On the other hand, a failed component may not be repaired properly because the cause of failure is not properly diagnosed; thus a delay will occur in returning the component to service. A mathematical model is developed which accounts for the two misclassification possibilities and a procedure is given which minimizes the misclassification probabilities, contingent on a fixed amount of available capital for this purpose. The model is of the Markov type with the simplifying assumption of constant transition rates (over time) among the various states.