How and When to Deploy Error Prone Sensors in Support of the Maintenance of Two-Phase Systems With Ageing

We consider the deployment of a sensor alongside a programme of planned maintenance interventions to enhance the reliability of two-phase systems. Such systems operate fault free until they enter a worn state which is a precursor to failure. The sensor is designed to report transitions into the worn state, but does so with error. The sensor can fail to report a transition when it occurs (false-negative), and can report one when none has taken place (false-positive). Key goals of our analyses are (i) the design of simple, cost effective schedules for the inspection, repair, and renewal of such systems, for use alongside the sensor; and (ii) the determination of the range of sensor operating characteristics for which the deployment of the sensor is cost beneficial. The latter is achieved via the computation of cost indifference curves which identify sensor operating characteristics for which we are indifferent to whether the sensor is deployed or not.