Risk based integrity modeling for the optimal maintenance strategies of offshore process components

Thodi, Premkumar (2011) Risk based integrity modeling for the optimal maintenance strategies of offshore process components. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (84MB)

Abstract

Ageing of components is a major threat to asset integrity in offshore process facilities. A robust maintenance strategy mitigates the effects of age-based structural degradations and reduces the threat of failure. Failure caused by structural degradations is a stochastic process. For maintenance strategies to be effective, the stochastic nature of failure has to be taken into consideration. Risk based integrity modeling (RBIM) is a newly-developed approach that aims at the protection of human life, financial investment, and the environment against the consequences of failure. RBIM quantifies the risk to which individual components are subjected and uses this as a basis for the design of a maintenance strategy. Risk is a combination of the probability and the consequence of failure. The major age-based structural degradations to be addressed include corrosion; such as uniform, pitting and erosion mechanisms; and cracking; such as stress corrosion, corrosion fatigue, and hydrogen induced cracking. In this study, component degradation processes are modeled stochastically to estimate the probability of failure using Bayesian analysis methods. Bayesian analysis improves the fidelity on the likelihood of future events by relating with the prior and posterior probabilities. Prior modeling is performed using judgmental studies and analyzing historic databases from similar installations. For the assessment of ageing assets and degradation mechanisms, field non-destructive test (NDT) data is used to establish the likelihood function. The posterior modeling is performed using a simulation-based Metropolis-Hastings algorithm and Laplace approximation since the prior-likelihood combinations are non-conjugate pairs. In this study, the consequences of failure are modeled using economic analysis to estimate the costs of failure, inspection and maintenance. The cost of failure includes lost production, loss of shutdown, cost of spill cleanup, loss caused by environmental damage and liability. The inspection and maintenance costs are estimated using the inspection and maintenance tasks, access, surface preparation, gauging defects, coating and restoration costs. Maintenance may be either minimal repair or replacement of components. The annual equivalent cost (AEC) of operation and maintaining a facility is the summation of the annual equivalent costs of failure, inspection, and maintenance. The cumulative posterior failure probability is combined with AEC to produce the operational life risk curve for a component. Since the risk curve is a convex function of the maintenance interval, then the optimum interval is the global minimum point. The operational risk is thus reduced to as low as reasonably practicable level by optimal maintenance.

Actions (login required)

View Item