Condition based maintenance in a plant management
program provides the ability to optimize the availability of
process machinery and greatly reduce the cost of maintenance.
Major improvements can be achieved in: maintenance costs,
unscheduled machine failures, repair downtime, spare parts
inventory, and both direct and in-direct overtime premiums. A
side benefit of condition based maintenance is the automatic
ability to monitor the mean-time-between-failures. This data
provides the means to determine the most cost effective time to
replace machinery rather than continue to absorb high
maintenance costs.
Many efforts have been made to develop methods and tools
to diagnose failures for predictive maintenance goal [2]-[5].
The essence of prognostics is the estimation of remaining life
in meaningful terms that would lead to a profound and
intelligent maintenance decision process. This in turn, would
lead to proactive maintenance processes minimizing downtime
of machinery and production and increasing of manufacturing
efficiency [6]. Prognostics are viewed as an add on capabilities
to diagnosis. They assess the current health of a system and
predict its remaining life, based on features that capture the
gradual degradation in the operational capabilities of a system
[7]. Prognostics are critical to improve safety, plan successful
missions, schedule maintenance, and reduce maintenance cost
and downtime [8].
Condition based maintenance techniques provide an
assessment of the system’s condition, based on data collected
from the system by continuous monitoring. The goal is to
determine the required maintenance plan prior to any predicted
failure. Therefore, the maintenance strategies aim to minimize
the cost by improvement of the operational safety and reduce
the severity and number of in-service system failures.
Accordingly to the ISO 13381-1:2004 standard, the activities
start with monitoring, followed by diagnostic, prediction and
posterior actions [9].
Various decision making approaches have been proposed
for maintenance strategy selection such as analytic hierarchy
process, fuzzy set theory, genetic algorithm, mathematical
programming, factor analysis, simple multi-attribute rating
technique, multi-criteria optimization, etc. [10]-[14].