In the first paper we focused mainly on establishing a sequential method for the selection of a maintenance technique for mechanical systems. The selection problem is studied for the components having considerable consequences and a non-negligible failure intensity. Identification of these components based on their failure mechanisms and failure causes is discussed and a modified version of the usual Failure Modes, Effects and Criticality Analysis, FMECA is suggested. A cumulative sum chart is proposed to identify a replacement time. This is explained by way of two examples. In order to run cost per unit time is proposed. The selection of a maintenance technique based on the intrinsic information of monitoring parameters is introduced.
Ability of a monitoring parameter in predicting failures, can be described as a probability of failure in the future conditionally on the current value of a model monitoring parameter. Thus, the second paper is focused on introducing a model relating a condition dependent failure rate to a component condition. For the selection of a cost-effective technique a modified proportional hazards model, PHM, is proposed. The modified model is able to reveal the effect of observing monitoring parameters and operating conditions on failure rate function.
A graphical method, based on a Generalized Total Time on Test, a GTTT- plot, for the selection of a cost-effective technique is proposed in the third paper. Also, a graphical method to estimate the threshold level for a cost-effective technique based on the same concept is suggested. These methods are illustrated by way of an example.
In this paper a sequential method for the selection of maintenance technique for mechanical systems is presented. It is based on technical knowledge concerning component failure mechanisms and failure causes. Classification of failures according to their consequences is stated. A modified version of failure modes, effects and criticality analysis is given and a way to identify significant components is discussed. Analysis of collected data is discussed and a method to identify a replacement level of a monitoring parameter is proposed. Criteria for the selection of a monitoring technique based on the intrinsic information of monitoring parameters are introduced. Finally, a decision rule for the selection of a suitable maintenance technique is suggested and a modified expression of the long run average cost per unit time is used as a distinguishing criterion between applicable strategies.
The cycle between two replacements based on condition monitoring is discussed. Models representing a condition dependent failure rate function reflecting the information in a monitoring parameter are suggested. For the selection of a cost-effective technique, a modified version of the Proportional Hazards Model, PHM, is proposed. Utilizing this modified model, it is possible to reveal the effect of monitoring parameters and operating conditions on the failure rate function. General comments on the application of the cost equation are given.
In this paper a graphical method for the selection of a cost-effective monitoring technique is suggested. Also, a graphical method to estimate the level at which components should be replaced when using a cost-effective technique, i.e. threshold level, is proposed. These two methods are based on the concept of the Total Time on Test, TTT-plot on age and condition dependent replacement is discussed.