Failure rate data sources

The distribution disk with this book includes a folder entitled BNLDATA which contains the file bnlgener.xls, a spreadsheet in EXCEL format. It is a collection of failure rate data drawn from many sources. The file refem.txt contains the references to the table s data. There are 1,311 data entries many cite different estimates by different organizations for the failure rate of the same... 

To properly use failure rate data, the engineer or risk analyst must have an understanding of failure rates, their origin and limitations. This chapter discusses the types and source of failure rate data, the failure model used in computations, the confidence, tolerance and uncertainties in the development of failure rates and taxonomies which can store the data and influence their derivation. 

Both of the sources above contain tWo types of failure rate data used in CPQRAs time-related failure rates and demand-related failure rates. Time-related failure rates, presented as failures per 10 hours, are for equipment that is normally functioning, for example, a running pump, or a temperature transmitter. Data are collected to reflect the number of equipment failures per operating hour or per calendar hour. 

Tolerance uncertainty arises from the physical and the environmental differences among members of differing equipment samples when failure rate data are aggregated to produce a final generic data set. Increasing the number of sources used to obtain the final data set will most likely increase the tolerance uncertainty. 

Failure rate data from public domain sources and derived from field failure-studies. Over 1,500 failure rates. 

The component failure rate data used as input to the fault tree model came from four basic sources plant records from Peach Bottom (a plant of similar design to Limerick), actual nuclear plant operating experience data as reported in LERs (to produce demand failure rates evaluated for pumps, diesels, and valves), General Electric BWR operating experience data on a wide variety of components (e.g., safety relief SRV valves, level sensors containment pressure sensors), and WASH-1400 assessed median values. 

SAIC provided much of the data used in this book from its proprietary files of previously analyzed and selected information. Since these data were primarily from the nuclear power industry, a literature search and industry survey described in Chapter 4 were conducted to locate other sources of data specific to the process equipment types in the CCPS Taxonomy. Candidate data resources identified through this effort were reviewed, and the appropriate ones were selected. Applicable failure rate data were extracted from them for the CCPS Generic Failure Rate Data Base. The resources that provided failure information are listed in Table 5.1 with data reference numbers used in the data tables to show where the data originated. 

Section 5.5 presents a data sheet for each cell in the taxonomy that contains failure rate data. Empty data cells are not presented. Filled data cells are listed by their CCPS Taxonomy number in Table 5.2 as an aid to the user. The CCPS data sheet format was developed from a number of sources including OREDA and IEEE Std. 500-1984. The format is presented in Figure 5.3, and its data elements are explained below ... 

Marti, H. F. and R. A. Waller. An Exploratory Comparison of Methods for Combining Failure Rate Data from Different Data Sources. Report No. LA-7556-MS, Los Alamos Scientific Laboratory, 1987. 

It is important to distinguish the types of time-related and demand-related equipment failure rate data that can be found and used in afire risk assessment. Basically, four types of data and corresponding data sources provide both time-related and demand-related failure data as shown in Table 6-3 and discussed in the following sections. 

Although facility-specific failure rate data on the specific equipment under study are preferred, often the only way to assemble sufficient data to satisfy study needs is to construct a "generic" data set. This data set is built using inputs from all of the facilities within a company, various facilities within the industry, literature sources, or commercial databases. 

In spite of their limitations, industry databases can be extremely valuable especially when no other data source exists. If the failure rate data is too high, the result will be a higher PFH/PFDavg. If this occurs and too much safety integrity is designed into a safety instrumented function, that is tolerable. 

One of the most popular failure rate databases is the OREDA database (Ref. 4). OREDA stands for "Offshore Reliability Data." This book presents detailed statistical analysis on many types of process equipment. Many engineers use it as a source of failure rate data to perform safety verification calculations. It is an excellent reference for all who do data analysis. 

The data is used in accordance with the guidance in NUREG/CR-2815 (Reference 5), and basic failure rate data is obtained from the EPRI ALWR Requirements Document (Reference 6) supplemented with data from the NREP Generic Data Base (Reference 7) and other nuclear sources. Maintenance and repair times are calculated as outlined in NUREG/CR-2815. The inspection and test times and frequencies are as specified in the System 80+ LCDs (See CESSAR-DC Chapter 16). 

A concern regarding the probabiiistic approach used in iEC 61508 and ANSI/ISA-84.00.01-2004-1 to determine the adequacy of the SiF design is that owners/operators could mistakenly assume unrealistically low failure rates for the SIF devices. The resulting erroneously low PFDavg could potentially lead to inadequate risk reduction. There are many sources of failure rate data, and sometimes it is difficult to decide what number best represents the device in the field application. ISA-TR84.00.02 provides more information on device failure rates, including a sampling of data from five owners/operators. 

For certification of the products, tools like FMEDA and Markov models are used. An FMEDA extends the FMEA techniques to include online diagnostic techniques and identify failure modes relevant to safety-instrumented system design. These PFD calculations may be carried out by commercial software tools (e.g., SILence by HIMA). In all these cases, bottom-up approaches were undertaken. Also it has been found tbat there has been a serious lack of reliable failure rate data and their variations among lab specifications, lab/field and field specifications. There are wide variations in the data obtained from these different sources. Therefore it is suggestive to use a band of data instead of discrete PFD and other failure rate data [5]. 

User s own valid failure rate data should be used within PFD calculations. Where this is not available use of appropriate recognised external data sources is acceptable. 

Quoted failure rates are therefore influenced by the way they are interpreted by an analyst and can span one or two orders of magnitude as a result of different combinations of the above factors. Prediction caiculations were explained in Chapter 5 and it will be seen that the relevance of failure rate data is more important than refinements in the statistics of the calculation. Data sources can at least be subdivided into site specific, industry specific, and generic and work has shown (Smith D J, 2000, Developments in the use of failure rate data...) that the more specific the data source the greater the confidence in the prediction. 

In the United States, Appendix III of the WASH 1400 study provided much of die data frequently referred to and includes failure rate ranges, event probabilities, human error rates, and some common cause information. The IEEE Standard IEEE 500 also contains failure rates and restoration times. In addition there is NUCLARR (Nuclear Computerized Library for Assessing Reliability) which is a PC-based package developed for the Nuclear Regulatory Commission and containing component failure rates and some human error data. Another US source is the NUREG publication. Some of the EPRI data are related to nuclear plant. In France, Electricity de France provides the EIReDA mechanical and electrical failure rate data base which is available for sale. In Sweden the TBook provides data on components in Nordic Nuclear Power Plants. 

Select appropriate failure rate data for the model(s) and justify the use of sources. 

Described in Chapter 6, a unique failure rate and failure mode data bank, based on over 50 published data sources together with Technis s own reliability data collection. FARADIP has been available for over 25 years and is now widely used as a data reference. It provides failure rate DATA RANGES for a nested hierarchy of items covering electrical, electronic, mechanical, pneumatic, instrumentation, and protective devices. Failure mode percentages are also provided. 

CCPS, 1989b, Process Equipment Reliability Data (Table 4.1-1) is a compilation of chemical and nuclear data. It assesses failure rates for 75 types of chemical process equipment. A taxonomic classification is established and data such as the mean, median, upper and lower (95% and 5%) values, source of information, failure by time and failure by demands are presented. 

Number and type of record The number of data points or tables of data presented in the resource or the number of events the data set reflects where available, the form in which the data are presented, such as failure rates or availability data, confidence intervals or error factors the raw data source used, sueh as surveys, plant records, tests, or judgment. 

Statistical Methods for Nonelectronic Reliability, Reliability Specifications, Special Application Methods for Reliability Prediction Part Failure Characteristics, and Reliability Demonstration Tests. Data is located in section 5.0 on Part Failure Characteristics. This section describes the results of the statistical analyses of failure data from more than 250 distinct nonelectronic parts collected from recent commercial and military projects. This data was collected in-house (from operations and maintenance reports) and from industry wide sources. Tables, alphabetized by part class/ part type, are presented for easy reference to part failure rates assuminng that the part lives are exponentially distributed (as in previous editions of this notebook, the majority of data available included total operating time, and total number of failures only). For parts for which the actual life times for each part under test were included in the database, further tables are presented which describe the results of testing the fit of the exponential and Weibull distributions. 

The failure rates and times-to-restore developed used a variety of data sources and data construction methodologies and are presented in Section 2. The principal methodology used is a kind of failure mode analysis for each component several principle modes of failure are analyed by characteristics including frequency of occurence, repair time, start-up time, and shut-down time. From these an average failure rate is developed and expressed as failures per million hours and mean time between failure(MTBF). 

Appendix III of this report provides a detailed description of the reliability data used in event tree and fault tree quantification. Because of its extensive operating experience and the uniqueness of the BRP design, BRP plant-specific data was used whenever possible. Plant-specific data sources included plant maintenance orders, control room log books, surveillance tests, LERs, event reports, deviation reports, plant review committee meeting minutes, and USNRC correspondence. The plant-specific data used spanned the period from 1970 to 1979. Data before 1970 did not include maintenance orders or surveillance tests and therefore were excluded. The plant-specific data collected for BRP is presented in detail in Appendix XIII. Table III-4 summarizes 30 plant-specific component failure rates and Table 11-06 contains plant-specific maintenance unavailabilities for 20 components. These tables are a summary of the BRP component failure and maintenance outages. 

WASH-1400 is a fundamental document for PRA methodology. The data appendixes contain a great deal of useful information on methods of data assessment. A large number of sources for data are considered, and very general failure rate estimates will produce only gross approximations. Since the advent of data collection schemes across and within plants, the WASH-1400 data are solely useful as a constituent to a data aggregation process or as widely bounded figures that provide a basis for comparison. 

Once it is determined that data exist, the next step is to begin the collection process. If sufficient thought and training is provided in the development and operation of the maintenance and operating reporting systems, much of the collection process can be automated. Automation assumes that a well-thought-out taxonomy is in place. If this is not the case, then an analyst must collect and review the records manually. In either case, the analyst must collect data from the plant sources previously discussed in order to determine the numerator (number of failures within a unique plant equipment population), and denominator (the operating time or number of demands for the equipment) of the equation to calculate failure rates. 

The information from these four sources of demands are then summarized and used for the failure rate calculations. Demand-related data sheets, appropriate to the hardware (Figures 6.6, 6.7, and 6.8) can be used to log information to compute demand-related failure rates. 

The most desirable source of equipment reliability data for a CPQRA is the operating experience of the process and plant being studied. Therefore, a chapter of this book provides information that will help an engineer locate raw plant reliability data and convert them to failure rates. However, the quality and confidenee level of the plant-specific data may be questionable because of operating and maintenance procedures, short relevant operating experience, and limited pieces of equipment available for study. The best data to use in a CPQRA are often a combination of generic and plant-specific data. 

Many companies have an internal expert who has studied these sources, as well as their own internal failure records, and maintains the company failure rate database. Some use failure data compilations found on the Internet. While the data in industry databases is not product specific or application specific, it does provide useful failure rate information for specific industries (nuclear, offshore, etc.) and a comparison of the data provides information about failure rates versus stress factors. 

Table 8-1 shows a comparison of data for a pressure transmitter. The failure rate numbers from the industry database sources are significantly higher than the FMEDA reports. 

Recently some analysis organizations have compiled comprehensive failure data source books and computer databases. The information is formatted to give failure rate as a function of failure mode. Often additional information about the product, such as Type A versus Type B, is provided. Some example pages are shown in Figures 8-1,8-2 and 8-3 (Ref. 9). 

---