AbstractProper maintenance of plant items is crucial for the safe and
profitable operation of process plants, The relevant maintenance
policies fall into the following four categories:
(i) preventivejopportunistic/breakdown replacement policies,
(ii) inspection/inspection-repair-replacernent policies,
(iii) restorative maintenance policies, and
(iv) condition based maintenance policies,
For correlating failure times of component equipnent and complete
systems, the Weibull failure distribution has been used, A new powerful method, SEQLIM, has been proposed for the estimation of the Weibull parameters; particularly, when maintenance records contain very few failures and many successful operation times.
When a system consists of a number of replaceable, ageing components, an opporturistic replacernent policy has been found to be cost-effective, A simple opportunistic rrodel has been developed.
Inspection models with various objective functions have been investigated, It was found that, on the assumption of a negative
exponential failure distribution, all models converge to the same
optimal inspection interval; provided the safety components are very
reliable and the demand rate is low, When deterioration becomes a
contributory factor to same failures, periodic inspections, calculated from above models, are too frequent, A case of safety trip systems has been studied,
A highly effective restorative maintenance policy can be developed if the performance of the equipment under this category can be related to some predictive modelling. A novel fouling model has been proposed to determine cleaning strategies of condensers,
Condition-based maintenance policies have been investigated. A simple gauge has been designed for condition monitoring of relief valve springs.
A typical case of an exothermic inert gas generation plant has
been studied, to demonstrate how various policies can be applied to
devise overall maintenance actions.
|Date of Award||Oct 1980|
- process plant
- Weibull distribution