Benefits of Risk Based Inspection Planning for Offshore Structures

The economical benefits of applying risk-based inspection planning (RBI) for offshore structures subject to fatigue are evaluated based on experiences from past industrial projects. To this end, the factors influencing the cost of inspection, repair and failure of structures are discussed and realistic values of these costs are presented. These are then applied to assess the expected costs from different inspection strategies, both riskbased strategies as well as inspection strategies with fixed inspection intervals for all potentially critical elements. By comparing these expected costs, the financial benefit of RBI is assessed. INTRODUCTION For offshore structures, risk and reliability based inspection planning (RBI) procedures have been developed and implemented since the 1980s, mostly for fatigue deterioration of fixed jacket steel structures, Skjong (1985), Madsen et al. (1989) and Fujita et al. (1989), but more recently also for ship and floating production storage and offloading systems subject to corrosion and fatigue, Lotsberg et al. (1999) and Goyet et al. (2004). While the significant computational efforts required by RBI hindered the applications in the past, this restriction has been resolved with the development of the generic approach to RBI, see Faber et al. (2000), Straub (2004), Faber et al. (2005) and Straub and Faber (2006), which facilitates the highly efficient application of RBI for portfolios of offshore structure. Although it has been observed that RBI in general reduces the amount of inspections considerably, Moan et al. (2000), the financial benefits of applying RBI strategies for offshore structures have not been systematically quantified in past publications. Furthermore, in the public domain little information is available on realistic estimates of costs related to the structural integrity management of these structures. As an example, in Dalane et al. (1990) the resulting expected cost for different designs and inspection strategies are compared for example details in fixed and floating offshore structures, yet without presenting the underlying cost model. In this paper, the factors influencing the costs of inspections and repairs are reviewed. Based on past experience, typical costs of different types of inspections and repair actions are presented for fixed and floating offshore steel structures, including FPSO’s, subject to fatigue damages. Furthermore, the factors influencing the cost of a failure in the structure are discussed. On this basis, the financial benefits of performing risk based inspection planning are determined for some typical cases by comparing the associated expected costs with the expected cost of an inspection strategy with fixed, predefined inspection intervals. This assessment is based on the generic approach to RBI. INSPECTION STRATEGIES The potential inspection strategies can be divided into three groups, namely prescriptive strategies, qualitative strategies and quantitative risk-based strategies. Prescriptive (or rule-based) inspection planning Prescriptive inspection plans require that inspections are performed for all joints in the structure at fixed intervals in time. Such rule-based inspection planning is still commonly 1 Copyright © 2006 by ASME applied in the offshore industry. Its main advantage is that it does not require further structural and risk analysis of the structure. Additionally, it may facilitate the planning of the logistic aspects of the inspection campaigns, because the number of inspections in each campaign can easily be adjusted to the available inspection capacities. Rule-based strategies are thus defined completely by the inspection interval . Insp T Δ Partly risk-based inspection strategies (qualitative or semi quantitative risk-based strategies) Some owners and operators of offshore structures implement an inspection policy which corresponds to a combination of the rule-based and the RBI approach, i.e., they apply inspection strategies which are partly risk-based. These strategies prescribe inspection intervals separately for groups of joints, in accordance with the considered qualitative and/or quantitative risk indicators. As an example, in Pemex (2000), inspection intervals are determined for fixed steel platforms as a function of various indicators, including the member importance (primary, secondary or tertiary member) and the calculated fatigue life. These partly risk-based inspection strategies are not considered explicitly in this paper. It is argued that although strategies based on qualitative indicators are preferable to purely prescriptive inspection planning strategies, they do have similar disadvantages. This holds in particular when considering fatigue deterioration, as qualitative indicators are poor in describing fatigue performance. Partly risk-based inspection strategies, which are based on quantitative indicators (such as the calculated fatigue life) are not treated in the following, as it is argued that with the availability of these indicators it is preferable to perform a fully quantitative RBI, as, once the indicators are available, the additional effort for such an analysis is small, as outlined in the following section. RBI strategies (quantitative risk-based strategies) RBI strategies are based on the priorization of inspection efforts in accordance with the risk reduction efficiency of the different alternative inspection actions. Ideally, such strategies should be based on the preposterior analysis of the Bayesian decision theory. In the following, the RBI model presented in Straub (2004) and Straub and Faber (2006) is summarized and subsequently applied for the numerical investigations; the practical application of similar models has been reported in, e.g., Pedersen et al (1992), Moan et al. (2000), Faber et al. (2005) or Chakrabarti et al. (2005). The deterioration mechanisms are represented by stochastic models of the defect size as a function of time, ( ) t S . For fatigue, ( ) t S is the crack depth and length as evaluated by a probabilistic fracture mechanics based model. Inspection qualities are commonly represented by a Probability of Detection (PoD) curve and a Probability of False Indication (PFI), which describe the likelihood of an inspection outcome given the state of the inspected component. Based on the PoD and PFI, structural reliability analysis or simulation techniques facilitate the updating of any deterioration model in the presence of an inspection outcome through the application of Bayes’ rule, as demonstrated by Madsen (1987). Through the assumption of no-indication at the inspections, the required inspection times to comply with a given threshold on the acceptable annual failure probability can be determined, as illustrated in Figure 1 for thresholds 10 yr and 10 yr. The assumption of no-indication implies that mitigation measures (repair, monitoring, follow-up inspections) are taken in case a defect is indicated at any of the inspections. The fact that inspection intervals increase with time reflects the increased confidence in the fatigue performance of the hot spot after the inspections. 0 10 20 30 40 50 60 70 80 90 100 10 10 10 Year t A n n u al p ro b ab ili ty o f fa ilu re Δp F