Modelling on-line reliability and risk to schedule the preventive maintenance of repairable assets in network utilities

This paper presents a methodology and a case study where the proportional hazards model is used to determine the reliability and risk of existing repairable systems in a distribution network utility. After introducing different issues relating to conditioning-maintenance management in these companies, we discuss the modelling possibilities to obtain on-line values for system reliability and risk. In this case, we try to model reliability considering the current operating time of the equipment, the number of maintenance interventions and the value of specific monitored parameters. Reliability is then used to calculate equipment risk for a given failure mode during a certain period, and in order to do so, the effect on the network of the failure mode is estimated from historical data. Finally, the paper presents a possible business process to schedule preventive maintenance activities according to previous findings and a case study.