OPTIMIZATION OF THE SYSTEM AVAILABILITY THROUGH CONDITION BASED MAINTENANCE

Abstract

This project deals with the optimization of the system availability through condition based maintenance using MARKOV analysis. The conventional techniques such as reliability block diagram, fault tree analysis and reliability graphs are of no use when there comes repairs and other dependencies. The MARKOV approach is considered for modeling as it can incorporate repair and other dependency features in the model. Generally, mechanical systems deteriorate gradually before they fail catastrophically. So it would be more appropriate, if multi state degradation is considered in the modeling; inspection is introduced in the model to monitor the condition and to take appropriate maintenance or repair action. Repair action depends upon the system state at particular interval. Here three modes of repair actions that are nominal repair, intermediate repair and major repair are taken. Repair rates can make system work in a prescribed manner, but cannot fix it always to initial stage or new state. If it is degraded to some lower state then it may be possible to recover it exactly, but at a highly degraded state it is almost impossible to fix it to a new state, therefore in such situations other repairs options like nominal repair just to make it work for some more time are taken. It has also noticed that by changing inspection time, availability varies. Therefore, the analysis was carried out to determine inspection interval when the availability becomes maximum using Markov approach.