Essay On Mechanical Maintenance

There are a large number of models for maintenance of mechanical components in the literature. Most of them assume that effect of maintenance interventions as “good as new”. Whether the action under gone is preventive or corrective, it is assumed that it is equivalent to its replacement. This assumption is questionable in many cases. Maintenance without any replacement leads to a significant level of greening of a system, either preventively or after repair action. Most of the time, the restoration of the performance is incomplete.

Every industrial installation is subjected to aging. The actual effect of wear-out depends on the system under consideration, but in every case aging reduces the global performance of the system, i.e. increase of the failure rate, larger energy consumption, dirt from nominal operating conditions, or reduction of the system productivity. It is quite different when comes to maintenance, for example, if the aim is to extend the intrinsic lifetime of piece of very expensive equipment to be systematic replaced. The objective of preventive interventions is to either reduce the effect of the system wear-out or delay the onset of these effects.

Deterministic optimization models have been proposed by various authors. Yao et al (2001) presented a model with two-layer hierarchical structure that optimizes the preventive maintenance scheduling for operations in semiconductor manufacturing industry. For the higher level, a planning model based on the Markov Decision Process framework captures both the cluster tools dynamic failure process and the production planning dynamics, is used to derive optimal policies for the long-term horizon. At lower level, an efficient mixed-integer programming model, which is in af...

... middle of paper ...

...e failure rate immediately preceding the failure. Their algorithm determines the optimal schedule of maintenance actions before each replacement action in order to minimize the total cost in a placing horizon.

We can find that most of the studies focus on single-component systems or simple and specific systems, which is not always applicable for real and general systems. In addition, not much work has been done in the area of age reduction and improvement factor models. Kamran S.Moghaddam and John S.Usher proposed preventive maintenance and replacement scheduling models that deal with multi-component system and can be applied to a wide variety of systems. Because they use the concept of age reduction and improvement factor in these models, they also developed mathematical and statistical models to estimate the improvement factor for imperfect maintenance activities.