Optimization of periodic maintenance using condition monitoring techniques and operational data

Abstract

This thesis contributes optimization of periodic maintenance using condition monitoring techniques and operational data so that maintenance can be performed before critical equipment fails. Periodic maintenance is carried out at regular intervals based on equipment’s failure history. In this case, the maintenance interval is not scheduled in terms of the actual deteriorating condition of equipment. Therefore, either over-maintenance or under-maintenance could be performed. This will cause overstocking of spareparts, high maintenance cost and unnecessary downtime By using Condition monitoring techniques such as vibration analysis, ultrasonic, oil analysis, infrared thermography, etc, the actual equipment condition can be monitored either continuously or periodically. By integrating the actual failure data acquired from condition monitoring techniques with historical (event) data, the current health condition and the remaining useful life of the equipment will be evaluated. Therefore, a suitable maintenance action can be scheduled and performed based on the current condition and projected remaining life of equipment. A procedure of condition monitoring is suggested in this thesis. Even though optimizing periodic maintenance using condition monitoring techniques leads to many benefits, there will be initial costs that may discourage the maintenance managers. This thesis presents a cost-benefit-analysis of condition based maintenance that helps the decision makers to decide the need of condition based maintenance to the given system. The last part of this thesis is to develop a methodology on how to select suitable equipment for condition based maintenance. Different criteria such as, criticality analysis, technical feasibility, economical analysis and others are considered for the selection process.