Khalifa, Mohamed (2012) Optimal risk-based inspection and maintenance (RBIM) planning for process assets. Doctoral (PhD) thesis, Memorial University of Newfoundland.
PDF (Migrated (PDF/A Conversion) from original format: (application/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.
Process assets are subjected to deterioration during operation. Inspection is carried out at pre-defined intervals and at prescribed locations. This monitoring strategy is needed to ensure that all assets perform their intended functions and that plant integrity is not threatened. Based on the outcomes of the inspection, a maintenance decision such as repair or replacement is made. -- This thesis developed a framework for optimal risk-based inspection and maintenance planning for process assets subjected to fatigue and corrosion. This framework includes two main parts: -- Inspection sampling: This part aims at estimating the required size of the inspection sample for assets/systems subjected to general corrosion and localized corrosion such as stress corrosion cracking, hydrogen induced cracking and pitting corrosion. -- Optimization of the risk-based inspection and maintenance (RBlM) plan: This part aims at determining the optimal inspection interval, inspection technique and maintenance activity (repair, replacement and/or alteration). -- In the first part of the framework, the required sample size is estimated to assss general and localized corrosion of process assets/systems. In the case of general corrosion, a Bayesian approach-based method is proposed for calculating the required sample size. The proposed method ensures that the error in the posterior estimate of the mean metal loss due to general corrosion does not exceed a pre-defined acceptable margin of error at a specified confidence level. An analytical formula to estimate the sample size is introduced. The sample size obtained using the proposed method is smaller than a sample size obtained using the classical method with the same confidence level. This reduces sampling inspection cost without affecting the precision of the estimate. -- In the case of localized corrosion, a different methodology is proposed to estimate the required sample size. The proposed methodology uses the extreme value and bootstrap methods. This methodology ensures that the predicted maximum localized corrosion using the extreme value method is within an acceptable margin of error at a specified confidence level. Two closed-form formulas are proposed for calculating the sample size in case of localized corrosion. The two formulas address the both situations when prior information is available or unavailable. A Bayesian updating approach is used to update prior information obtained from previously performed inspections and engineering judgement. -- In the second part of the framework, a methodology for the optimal selection of an RBIM plan is proposed. This methodology is comprised of the following main steps: classification of asset's components/areas according to criticality of deterioration, asset deterioration modeling, risk assessment, cost estimation (inspection and maintenance) and finally the selection of optimal inspection intervals and a maintenance strategy. To solve the optimization problem, an objective function is formulated as a function of the present value of inspection cost, repair/replacement cost, risk of failure and the remaining value of the asset after a specified period of time. The selection of the optimum inspection interval, inspection technique and maintenance activity is based on minimizing the objective function subject to a safety constraint that the risk of failure over the lifetime of the asset does not exceed an acceptable level. The proposed methodology allows for a minimization of the inspection and maintenance cost over the lifetime of a deteriorated asset/system without compromising the safety. -- The developed RBIM framework will help operators to make well informed decisions, which will result in cost effective asset integrity management and a higher level of safety.
|Item Type:||Thesis (Doctoral (PhD))|
|Additional Information:||Includes bibliographical references (leaves 153-166).|
|Department(s):||Engineering and Applied Science, Faculty of|
|Library of Congress Subject Heading:||Plant maintenance--Management--Mathematical models; Reliability (Engineering)--Mathematical models; Industrial equipment--Maintenance and repair; Industrial equipment--Inspection; Bayesian statistical decision theory|
Actions (login required)