Common Cause Failure Events Research Articles

Methods for Estimating Parameters of the Common Cause Failure Model Based on Data with Uncertainty

In this paper, we propose a new method to deal with uncertain data in the context of Common Cause Failure (CCF) analysis. Uncertain CCF data refer to the data for which the number of components involved in the failure events is not exactly known. We introduce a new formalism to describe uncertain CCF data to avoid subjective probabilities for the number of failed components in each CCF event that are used in classical methods such as the impact vector method. The parameters of the [Formula: see text]-factor model are estimated using the maximum likelihood method relying on properties of the nested Dirichlet distribution and grouped Dirichlet distribution. A data augmentation technique with an expectation-maximization algorithm is also developed for some schemes of data with uncertainty. Finally, we evaluate the performance of the proposed method through numerical simulations and illustrate its application using an example from the literature.

Assessing the Impact of Common Cause Failures on Site Risk within Level 1 Multi-Unit PSA

Common cause failures (CCFs) may lead to the simultaneous unavailability or failure of numerous components in the nuclear power plant because of the existence of a shared cause when an initiating event disrupts the normal functioning of nuclear power plants. The presence of common cause failures (intra-unit and inter-unit) can be recognized in a multi-unit probabilistic safety assessment (MUPSA) as a crucial dependency factor that can influence accident scenarios and the core damage frequency (CDF), as CCF may affect the availability and proper operation of mitigating systems. Since such failures are likely to significantly undermine the benefits of the concept of redundancy in nuclear power plant systems, it is necessary to identify the CCFs that contribute to the core damage in a multi-unit site and analyse their overall quantitative magnitude and qualitative proportions. In this study, a twin-unit generic pressurized water reactor (PWR) nuclear plant is modeled using the AIMS-PSA software. For the loss-of-offsite-power (LOOP) and station blackout (SBO) events, the site CDF was calculated, and the cut-sets produced by this quantification were examined for the modeled CCF basic events in the fault trees. The quantitative and qualitative contributions of the CCFs to the frequency of site core damage were examined. CCFs in the modeled fault trees contributed to 4.58% to the site CDF of the combined LOOP followed by SBO event. In the LOOP event alone that leads to core damage, the CCF contributed 4.58% to the site CDF while CCFs contributed 17.19% to the site CDF in the SBO event alone that leads to core damage. With CCF events considered in the modeling process, the site CDF estimated with CCF events increased by 7.53% in the combined LOOP followed by SBO event. In the LOOP event alone that leads to core damage, inclusion of CCF events in the modeling increased the site CDF by 7.42%. A 15.66% increase in site CDF was recorded in the SBO event alone that leads to core damage as compared to modeling without CCF events. The results show how crucial the common cause failure contribution is to site CDF. The safety of the nuclear plant at a site is impacted by an increase in site CDF when common cause failures are considered. The various CCF fundamental event compositions and their percentage contributions were explicitly examined by the minimal cut-sets which leads to core damage in the units. In conclusion, this study’s findings can help us better understand how CCFs increase multi-unit site risk and can also act as a starting point for future studies on the qualitative and quantitative categorizations of CCF effects within MUPSA.

A systematic approach to estimate an inter-unit common-cause failure probability

Inter-unit common-cause failure (CCF) analysis is one of the technical issues for a probabilistic risk assessment for multiple nuclear reactors in a site. It is because inter-unit CCF events are rare and the symmetric assumption in intraunit CCF analysis is no longer valid. To overcome these difficulties, we propose a systematic approach to estimate the inter-unit CCF probability using conventional CCF databases and engineering judgments. The proposed approach explicitly considers two sources of asymmetry in inter-unit CCF: the commonalities of common-cause coupling factors and surveillance tests. The proposed approach starts from a symmetric CCF probability of components under a nonstaggered test scheme and extends it to take the asymmetry into account. The proposed approach is applied to the CCF analysis of the hypothetical case of three nonidentical emergency diesel generators and is compared with the conventional CCF analysis. The proposed approach can reduce the conservativeness in the inter-unit CCF probabilities estimated by the conventional CCF analysis and improve the quality of multiunit probabilistic risk assessment (PRA), where inter-unit CCF events significantly contribute to the site risk. The proposed approach demonstrates its applicability to an inter-unit CCF analysis where the asymmetries play an important role in inter-unit CCF probability.

Limitations of traditional tools for beyond design basis external hazard PRA

Probabilistic risk assessment (PRA) is being used increasingly by the nuclear industry for safety during normal operations as well as for the protection against external hazards. Computation of total risk in an external hazard PRA is dependent on hazard assessment, fragility assessment, and systems analysis. A systems analysis for propagation of component fragilities is conducted using event and fault trees. The event and fault trees for an actual power plant can be fairly large in size, which imposes computational challenges. Hence, certain assumptions are employed for computational efficiency. These assumptions typically represent the conditions imposed during the design basis (DB) scenario. The traditional PRA tools based on these assumptions are also widely applied to perform risk assessment in the context of beyond design basis (BDB) scenarios. However, some of these assumptions may not be valid for certain BDB scenarios. In addition, the probability of dependent failures also increases in BDB scenarios due to common cause failures (CCF) which usually results from design modifications, human errors, etc. In this manuscript, a simple and a relatively more complex illustrative examples are used to show the limitation of these assumptions in the numerical quantification of risk for the case of BDB conditions. Case studies with CCF events across multiple fault trees are also presented to illustrate the effect of these assumptions when traditional approach is used in BDB risk assessment. It is shown that the assumptions are valid for the case of DB conditions but may lead to excessively conservative risk estimates in the case of BDB conditions. A Bayesian network based top-down algorithm is proposed as an alternative tool for accurate numerical quantification of total risk in systems analysis.

A development of methodology for assessing the inter-unit common cause failure in multi-unit PSA model

The inter-unit common cause failure (CCF) is an event that multiple components are failed simultaneously by common factors may existent in nuclear power plants on a site. The inter-unit CCF can be identified as one of the inter-unit dependencies that has an impact on the multi-unit probabilistic safety assessment (MUPSA). For instance, it may adversely affect the accident mitigation in accident scenarios of multiple units. Therefore, the MUPSA model should be developed considering the inter-unit CCF.In this study, we proposed two methods with the Swain's dependency model and the CCF data in the single-unit PSA (SUPSA) to calculate the inter-unit CCF. The Swain's dependency model is used to calculate the conditional probability considering the dependency level between each CCF event in two different units. To perform the case study, the inter-unit CCF events of emergency diesel generators (EDGs) and alternate AC diesel generators (AAC DGs) in 4 units of different pressurized water reactor (PWR) types were analysed in the case of the multi-unit loss of off-site power (MULOOP). It is expected that methodologies in this study will contribute to the multi-unit level 1 PSA model development as well as evaluating the site risk in the future.

Reliability and importance analysis of uncertain system with common cause failures based on survival signature

Redundant design has become the commonly used technique for ensuring the reliability of complex systems, which calls for great concern to common cause failure problems in such systems. Incomplete data in combination with vague judgments from experts introduce imprecision and epistemic uncertainties in the performance characterization of components. These issues need to be taken into account for assessing the system reliability. In this paper, a comprehensive reliability assessment method is presented by adopting the concept of survival signature to estimate the reliability of complex systems with multiple types of components. Particular attention is devoted to common cause failures (CCFs), which are modeled and quantified by decomposed partial α-decomposition method. Uncertainties caused by incomplete data for CCF events are reduced by hierarchical Bayesian inference. The component importance measure is enhanced to assess the importance of various possible CCF scenarios and to identify their potential impact on system reliability. The presented method is used to analyze the reliability of a dual-axis pointing mechanism for communication satellite, which is a commonly used satellite antenna control mechanism. The engineering application demonstrates the effectiveness of the method.

Bayesian network-based seismic damage estimation for power and potable water supply systems

As critical infrastructures being more interdependent with each other and vulnerable to natural disasters, assessment of cascading failure in the coupled system becomes a significant task for ensuring reliability. Above all, the system interruptions caused by power outages have been paying attention to, and extensive research efforts have explored for the last decades. However, current studies tend to focus on system-level interdependency, which makes it hard to identify which component plays a critical role in the performance of infrastructure. In this context, this paper provides a method to estimate the service disruptions of power and potable water supply system after an earthquake. Two types of failures, common-cause and cascading, are modeled as a Bayesian network, which includes (a) an earthquake, (b) common-cause failure events, (c) cascading failure events, and (d) final service disruption. To quantify the probability of service disruptions under varying earthquake scenarios, a modified seismic ground motion prediction equation, fragility functions, and concept of functionality are used. The proposed Bayesian model is tested through case simulations run on data collected from Daegu City, South Korea. By presenting a total of 15 types of multi-state event nodes and all possible damage propagation causing disturbance of power and potable water supply systems, this paper gives knowledge for understanding the resilience of infrastructures in network perspectives.

Fuzzy Reliability Assessment of Safety Instrumented Systems Accounting for Common Cause Failure

Reliability of safety instrumented systems (SISs) is a critical measure to ensure production safety of many industries. This paper focus on low-demand SISs. The reliability of these SISs is quantified by evaluating their probability of failure on demand (PFD). However, due to lack of knowledge, and/or vague judgments from experts, epistemic uncertainty associated with the parameters of components' degradation models is inevitable. Meanwhile, common cause failure (CCF) of two or more components caused by shared environments, often exists in SISs, reliability assessment for a SIS, therefore, becomes a challenging task. In this paper, fuzzy reliability assessment for SISs is conducted by taking account of the CCF among components of a SIS. The fuzzy Markov model is utilized to characterize the degradation process of components under fuzzy environment. The fuzzy state probability distribution of components is, then, calculated by formulating a set of constrained optimization models. Based on the fault tree of a SIS, the fuzzy PFD of the entire SIS with CCFs is formulated by using the $\beta $ -factor to quantify the CCFs. The fuzzy PFD at any $\alpha $ -cut level is, therefore, computed by a constrained optimization model. According to the optimization of parameters of SIS, the lower bound and upper bound of fuzzy PFD of SIS can be determined. Finally, we can quantify the effectiveness of CCF event for assessing the fuzzy PFD of SIS.

Extend GO Methodology to Support Common-Cause Failures Modeling Explicitly by Means of Bayesian Networks

As a success-oriented system reliability and safety-analysis technique, the GO methodology has been applied in a variety of real-world safety-critical industrial fields. Common-cause failure (CCF) is the simultaneous failure of multicomponents within a system due to the same root cause. An enhancement approach for the original GO methodology is proposed in this paper to support CCF modeling and calculation both in graphical modeling aspect and algorithm aspect. First, a new concise and formalized GO operator (named CCO) is introduced to represent complicated CCF event, which makes the CCF modeling process intuitive and concise for analysts. In algorithm aspect, the mapping rule is given and demonstrated to transform new CCO operator with impacted multiple operators to the corresponding Bayesian network (BN) fragment. Second, general mapping programmable process is presented on transforming any CCF enhanced GO model to the corresponding BN. Furthermore, using BN's inference capability, the enhanced GO model with CCF can be calculated efficiently. Nevertheless, the diagnosis process can be performed to investigate the weak points of the modeled system. Finally, a case study is performed to demonstrate the modeling process by means of CCF enhanced GO model. The calculation result shows that CCF has significant influence on the system reliability. Using diagnosis analysis, the CCF event can be confirmed as the major cause leading to system failure.

Recent Insights from the International Common-Cause Failure Data Exchange Project

Abstract Common-cause failure (CCF) events can significantly impact the availability of safety systems of nuclear power plants. For this reason, the International Common Cause Data Exchange (ICDE) project was initiated by several countries in 1994. Since 1997 it has been operated within the Organisation for Economic Co-operation and Development (OECD)/Nuclear Energy Agency (NEA) framework and has successfully been operated over six consecutive terms (the current term being 2015–2017). The ICDE project allows multiple countries to collaborate and exchange CCF data to enhance the quality of risk analyses, which include CCF modeling. As CCF events are typically rare, most countries do not experience enough CCF events to perform meaningful analyses. Data combined from several countries, however, have yielded sufficient data for more rigorous analyses. The ICDE project has meanwhile published 11 reports on the collection and analysis of CCF events of specific component types (centrifugal pumps, emergency diesel generators, motor operated valves, safety and relief valves, check valves, circuit breakers, level measurement, control rod drive assemblies, and heat exchangers) and two topical reports. This paper presents recent activities and lessons learnt from the data collection and the results of topical analysis on emergency diesel generator CCF impacting entire exposed population.

Unification of Common Cause Failures’ Parametric Models Using a Generic Markovian Model

The main objective of this paper is to provide a unified Markov model which could implement any parametric models devoted to the treatment of common cause failures (CCFs), namely: beta factor, multiple greek letter, alpha factor, multiple beta factor, and binomial failure rate. The choice of the Markovian representation is motivated by the fact that classical reliability approaches (e.g., Fault trees) are not able to catch the dynamic aspect induced by CCF events. The proposed Markovian model is also capable of dealing with any M-out-of-N configuration. It is illustrated on the basis of a system made up of four identical components (N = 4) in order to quantitatively examine the differences between the first four mentioned parametric models. For that purpose, the considered performance indicators are the average unavailability (Uavg) and the average unconditional failure intensity (i.e., failure frequency) (wavg).

A pragmatic approach to estimate alpha factors for common cause failure analysis

Most of the modern technological systems are deployed with high redundancy but still they fail mainly on account of common cause failures (CCF). Various models such as Beta Factor, Multiple Greek Letter, Binomial Failure Rate and Alpha Factor exists for estimation of risk from common cause failures. Amongst all, alpha factor model is considered most suitable for high redundant systems as it arrives at common cause failure probabilities from a set of ratios of failures and the total component failure probability QT. In the present study, alpha factor model is applied for the assessment of CCF of safety systems deployed at two nuclear power plants. A method to overcome the difficulties in estimation of the coefficients viz., alpha factors in the model, importance of deriving plant specific alpha factors and sensitivity of common cause contribution to the total system failure probability with respect to hazard imposed by various CCF events is highlighted. An approach described in NUREG/CR-5500 is extended in this study to provide more explicit guidance for a statistical approach to derive plant specific coefficients for CCF analysis especially for high redundant systems. The procedure is expected to aid regulators for independent safety assessment.

Estimation of common cause failure parameters for essential service water system pump using the CAFE-PSA

This paper presents the estimation results of the common cause failure (CCF) parameters of essential service water system (ESWS) pump failure to run for KX nuclear power plant (NPP) in Korea. Until now, the generic values of the CCF parameters have been mainly used in most probabilistic safety assessment (PSA) projects for the Korean NPPs. The PSA results for KX NPP showed that the CCF events of ESWS pump failure to run was identified as one of dominant contributors to its internal event core damage frequency (CDF). Thus, we performed the plant specific detailed CCF analysis to estimate CCF parameters of ESWS pump failure to run for KX NPP with the CAFE-PSA, a program to analyze CCF events in the ICDE database. Reasonable values of CCF parameters were obtained through performing plant specific detailed CCF analysis. The estimated Alpha Factor with three out of three failure criterion was about one half of that for recent US NRC CCF parameters. The re-quantification results on the CDF of KX NPP with the new estimated Alpha Factor showed that originally estimated CDF with generic Alpha Factor decreased by 16.84% and the contribution of the sum of cutsets for the CCF events of ESWS pump failure to run to internal event CDF decreased from 20% to 3.29%.

Bayes geometric scaling model for common cause failure rates

This paper proposes a mathematical model to associate key operational, managerial and design characteristics of a system with the system's susceptibility towards common cause failure (CCF) events. The model, referred to as the geometric scaling (GS) model, is a mathematical form that allows us to investigate the effect of possible system modifications on risk. As such, the presented methodology results in a CCF model with a strong decision-making character. Based on a Bayesian framework, the GS model allows for the representation of epistemic uncertainty, the update of prior uncertainty in the light of operational data and the coherent use of observations coming from different systems. From a CCF perspective these are particularly useful model features, because CCF events are rare; hence, the operational data available is sparse and is characterised by considerable uncertainty, with databases typically containing events from nominally identical systems from different plants. The GS model also possesses an attractive modelling feature because it significantly decreases the amount of information elicited from experts required for quantification.

Insights from analysing the Nuclear Regulatory Commission's common-cause failure database

This paper summarizes key insights from studies of common-cause failures (CCFs) of emergency diesel generators, motor-operated valves, motor-driven pumps, and circuit breakers from 1980 to 2005. The data studied were derived from the Nuclear Regulatory Commission's CCF database, which is based on US commercial nuclear power plant event data. The insights are the result of an in-depth review of the CCF data. Trends of the annual number of CCF events show that the number of events has been decreasing over time. The insights can help to focus inspection and utility maintenance activities.

Approximate formulas for treating asymmetrical common cause failure events

A parameter estimation of common cause failure (CCF) events for parametric models such as the alpha factor and Multiple Greek letter methods is based on a symmetry assumption that the probabilities of CCF events involving similar components are the same. However, many components in nuclear power plants (NPPs) have asymmetrical characteristics for their failures. In this paper, we derived the approximate formulas for modeling asymmetrical CCF events and estimating their parameters by using a decomposition approach. The total failure events of a component including the CCF events were divided into their symmetry part and their asymmetrical part. Under the assumption that the CCF events of each divided part were the symmetrical CCF events within them, the alpha factor method and the basic parameter method were employed for a parameter estimation of the asymmetrical CCF events. The derived approximate formulas were applied to the CCF analysis of three emergency diesel generators (EDGs) to demonstrate their adequacy.

Coupling Model: A Common-Cause-Failure Model with Consideration of Interpretation Uncertainties

The quantification of common-cause failures (CCFs) is often connected with uncertainties in how to interpret observed CCF events and with how far they are applicable to the specific group of components in question. A method has been developed that allows consideration of these kinds of uncertainties on the basis of a modification of the Binomial-Failure-Rate model. The quantification of interpretation uncertainties by means of interpretation alternatives is discussed as well as their effects on the estimation of the coupling parameter of the underlying CCF model. The estimation of the coupling parameter under consideration of the aforementioned uncertainties is performed by a Bayesian approach. To facilitate the specification of interpretation uncertainties, a default proposal of the interpretation vector is automatically generated on the basis of component fault states gained by expert judgment. Modification of the default vector is possible depending on engineering judgment of technical or operational differences between the observed and the target group of components.

ON IMPORTANCE MEASURES USED IN COMMON CAUSE FAILURE QUANTIFICATION

In quantitative analysis of common cause failures (CCFs), importance measures for ranking the importance of the CCF events will be significant to limit the CCF analysis framework, find out efficient defence strategies against CCF, and make sensitivity analysis for them. In this paper, three importance measures are defined for different types of common cause failure events: Structure Importance, Probability Importance, and β-Importance. Simplified algorithms for calculating these importance measures are developed.