OREDA Offshore Reliability-Data Handbook 4th Edition Published by: | OREDA Participants Prepared by: SINTEF Industrial Management Distributed by: | Det Norske Veritas (DNV)OREDA-2002 2 OREDA-2002 Copyright © 2002 by the OREDA® companies! ENIS.p.A/AGIP Exploration & Production BP Exploration Operating Company Lid ExxonMobil International Ltd. Norsk Hydro ASA Phillips Petroleum Company Norway Statoil ASA Shell Exploration & Production TotalFinaElf All rights reserved. No part of this publication may be reproduced, stored in any retrieval system or transmitted in any form or by any means, electronic, mechanical photocopying, recording or otherwise, without prior written permission from the copyright holders. Comments or questions to the book can be directed to: SINTEF Industrial Management Safety and Reliability OREDA Project Manager NO-7465 Trondheim NORWAY Phone: +47 73 592756 Telefax: +47 73 59 28 96 ewmail: oreda@sintef.no http: —_http:/Avww.sintef-noforeda/handbook/ Copies of the book may be ordered from: Det Norske Veritas NO-1322 Hevik NORWAY Phone: +47 67 $799 00 Telefirc: +47 67 57 74.74 e-mail: oreda@dnv.com up: —_http://www-dnv.com/ogpiforeda, esreda/oreda/oreda.htm ISBN 82-14-02705-5 OREDA® = registered trade mark name of OREDA (Offshore Reliability Data) ' The companies listed below are those being members of OREDA in 2002. @OREDAOREDA-2002 3 OREDA-2002 DISCLAMER Information in this document is intended to permit data utilisation in the interests of evaluating and improving safety and reliability in the oil & gas industries (exploration & production). Publication of said information does not imply verification or endorsement of the information. Neither the publishers nor the originators of data included in this document assume any liability to parties adopting any product, process, or practice based upon the usage of the information. ‘The correctness of the data listed in this document is assured only to the limits that the producers of each individual chapter, and their sources, can ensure correctness. ‘A listing, description, or definition of any item in this document is not to be regarded, by implication or otherwise, as in any manner licensing any person or corporation, or conveying any rights or permission to manufacture, use, or sell any device or item that may in any way be related thereto. This document may not be employed in any manner that might be construed as endorsement or censue of any itcm listed herein, and should only be used within the scope of the intent stated above. The existence of this document does not imply that there are no other ways to present the offshore reliability data compiled in this book. Furthermore, the viewpoint or technical perspective expressed in this publication is subject to changes brought about by technological developments and comments received from users of the former OREDA Reliability Handbooks (-84, -92, -97). Requests for interpretation may arise regarding the meaning of parts of the handbook as they relate to specific applications. Those requests should be directed to SINTEF Industrial ‘Management at the addresses given. As publications issued by the OREDA project represent a consensus of all participating companies, it is important to ensure that any interpretation has also received the joint agreement of those companies. For this reason, it may in some cases be necessary to consult the OREDA Steering Committee. In these cases it will not be possible to provide an instant response to interpretation requests, except in those cases where the matter bas previously received formal consideration, OREDAOREDA-2002 4 OREDA-2002 PREFACE The reliability, availability, maintenance and safety (RAMS) of offshore exploration and production ( & P) facilities are of considerable coneem to employees, companies and authorities; RAMS analyses ate ‘carried out to provide a basis for decisions in offshore engineering, fabrication and operations. Ti order to allow these analyses to be conducted, a source of reliability data is required. The OREDA participating companies have responded to this need by publishing four reliability data handbooks. The previous editions of the OREDA handbook were published in 1984, 1992 and 1997 respectively. The current 2002 edition has not been changed significantly compared to the former -97 edition, However, it should be noted that drivers are no longer included within the boundary of the driven unit but defined as separate entities. As the 2002 edition contains a lot more subsea data than previously, these data are presented in a fashion more adapted to subsca application. The handhook is divided into twa parts. Part I describe the OREDA project, different data collection phases and the estimation procedures used to generate the handbook. In Part II the reliability data tables are presentedfor both topside and subsea equipment. In order to interpret and apply the data given in Part II in the most correct manner, it is recommended to read Part I before using data shown in Part II. While each new edition of the handbook is considered to be a step forward, it is realised that the quality and the quantity of the data presented in the handbook may be inadequate for certain applications. The OREDA Steering Committee's intention is therefore to keep the handbook up to date according to the state of technology by publishing new editions. ‘The current handbook covers phase IV and V of OREDA data collection. This represent broadly the time period of 1993 — 00. It may, however, be seen that some data collected may be beyond this time period. Data collected are in general not covering the whole lifetime of equipment, but typically a time windows of 2 — 4 years of operation. Infant mortality failures are as a rule not collected for topside equipment, hence, the data collected are from the normal steady-state operating time period as illustrated in Figure 5. For subsea equipment failures are in general collected on a whole lifetime basis, i.e. including the infant mortality period. ‘The source data for this book are stored in a computer database except for phase I. The database and the associated software are available only to the oil companies participating in the OREDA project. The data presented in this handbook are extracted and compiled from this database and presented as generic data tables. The database does, however, contain additional information -and information on a more detailed level- than what is covered in this handbook. For those who want to exploit this possibility, one of the participating oil companies needs to be contacted. A list with contact names for each participating company can be found in the OREDA Homepage: http:/OREDA.com @OREDAOREDA-2002 5 OREDA-2002 CONTENTS DISCLAMER.... PREFACE PARTI. INTRODUCTION THE OREDA PROvECT.. PROJECT PHASES. . PARTICIPANTS ORGANISATION... EQuIpMENT CATEGORIES COVERED IN THE Scope OF THE OREDA HANDBOOK... Limrranions: ‘THE OREDA TOPSIDE DATA STRUCTURE. SYSTEM HIERARCHY. Equipment BOUNDARIES .. Inventory Data. Faivure Data... ‘THE OREDA SUBSEA DATA STRUCTURE. ‘SySTEM HIERARCH’ Equipment BOuNDaR! Inventory DATA... FAILURE EVENT AND MAINTENANCE DATA, ESTIMATION PROCEDURES. FAILURE RATE... ESTIMATORS AND UNCERTAINTY Liwits ror A HOMOGENEOUS SAMPL MuLTSaMPLE PROBLEMS... cn ESTIMATION OF DEMAND PROBABILITIES. TOPSIDE DATA TABLE FORMATS... Data TABLE, RELIABILITY DATA.... DATA TABLE, MAINTAINABLE ITEM VERSUS FAILURE MODE, DATA TABLE, FAILURE DESCRIPTOR VERSUS FAILURE MODE SUBSEA DATA TABLE FORMATS... DATA TABLE, RELIABILITY DATA... Data TABLE, COMPONENT VERSUS FAILURE MODE... Data TaBLe, SUBUNIT VERSUS FAILURE Move. @OREDA: : OREDA-2002 6 DATA TABLE, EQUIPMENT UNIT VERSUS FAILURE MOOE .. DATA TABLE, FAILURE DESCRIPTOR VERSUS FAILURE MODE... MISCELLANEOUS ESTIMATION PROCEDURES... No FAILURES ARE OBSERVED FOR A SPECIFIC FAILURE MODE .. WEIGHTING OREDA-O2 Data witd OrHeR DATA SOURCES. DEFINITIONS .... REFERENCES .. £ 8 & 68 8 88 PART Il RELIABILITY DATA PRESENTATION.. MACHINERY .. Compressors. Ges Turbines Pumps... : Combustion Engines 233 Turboexpanders. 270 ELECTRIC EQUIPMENT.. 291 Electric Generators. 201 Electric Motors 332 MECHANICAL EQUIPMENT. 373 Heat Exchangers .. 373 Vessels. 414 Heaters and Boilers 487 CONTROL AND SAFETY EQUIPMENT. “Bt Fire and Gas Detectors. .51t Process Sensors. 536 Valves... 566 ‘SUBSEA EQUIPMENT 804 ‘Common components Control systems Manifold Flowline....... Subsea isolation system 822 Risers 824 Running tool. 827 Wellhead and X-mas tre 830 @OREDAOREDA-2002 7 OREDA- 2002 PARTI ‘© OREDAOREDA-2002 8 OREDA-2002 INTRODUCTION THE OREDA PROJECT The Offshore Reliability Data (OREDA) project was established in 1981 in co-operation with the Norwegian Petroleum Directorate. The initial objective of OREDA was to collect reliability data for safety equipment. The current organisation, as a co-operating group of several oil companies, was established in 1983, and at the same time the scope of OREDA was extended to cover reliability data from a wide range of equipment used in oil and gas exploration and production. Offshore topside and subsea equipment are primarily covered, but some onshore E & P equipment is also included. The main objective of the OREDA project is to contribute to an improved safety and cost- effectiveness in design and operation of oil and gas exploration and production facilities; through collection and analysis of maintenance and operational data, establishment of a high quality reliability database, and exchange of reliability, availability, maintenance and safety (RAMS) technology among the participating companies. PROJECT PHASES Phase I (1983 - 1985) ‘The purpose of phase I was to collect and compile data from ofishore drilling and production operations, The data were published in the OREDA-84 handbook. An objective of the handbook was to demonstrate the ability of the eight participating oil companies to co-operale on this issue and create a forum for a common co-operative process in this field. Data was collected on a wide area of equipment (large population) but not with as much detailed information as in later phases. Phase IT (1987 - 1990) The scope was adjusted to only collect data on production critical equipment, to improve the quality of the data, and to store the data in a PC database format, A tailor-made PC program (calied the OREDA software) was developed to aid the collection and analysis of the data. The data were published in the OREDA-92 handbook. This Handbook also contains the data collected in phase 1 Phase [IT (1999 - 1992) ‘The number of equipment categories was increased, and more data on maintenance programs were collected. The data quality was improved by means of the comprehensive “Guidelines for Data Collection” and through quality control. The OREDA software was modified into a more general-purpose data collection tool, and its user interface was improved. The data collected in this phase are contained in the OREDA-97 handbook, Phase IV (1993 - 1996) A new general software was developed for data collection and analysis, plus specific software and procedures for automatic data import and conversion. Data were collected mainly for the @OREDAOREDA-2002 9 QREDA-2002 same equipment as in phase Til, and the data collection was - to a greater extent - carried out by the companies themselves. Data on planned maintenance are included. Phase V (1997 - 2000) Some new equipment classes were included and more focus was given on collecting subsea data. As a parallel activity, the ISO standard 14 224: “Petroleum and natural gas industries - Collection and exchange of reliability and maintenance data for equipment" was developed and issued in July 1999. A revised version including downstream equipment is currently being developed by ISO TC67 Workgroup 4 with the secretariate at NTS in Oslo. (See: http:/www.nts.no/) Phase VI were completed in 2001 and phase V1 is planned to last 2002 - 2003. Lip-to-date information on the OREDA project is available on the Internet address: http:/Avww.creda.com PARTICIPANTS During phase IV and V several changes have been expcricnced in company participation in OREDA as to new companies joining and leaving OREDA as well as companies being merged/sold. The following summarise the companies that have contributed with data in these phases: ‘Companies Phase iV_| Phase V | Gommonte AGP. y BP. a v hewn y z Marged wilt Texaco ELF z v ‘Matged with TOTAL Exsofoon v T “(wlosged wih Mobil) Norsk Hydro a x Phillips Petroloum Company Nonway_ Y Statoil Y x ‘Saga Pelroleum Y ‘Sold to Norsk Hydro Shell v T Texaco vo Merged with Chevron TOTAL ni Merged with Elf ORGANISATION OREDA is managed by a Steering Committee with one member and one deputy member from each of the participating oil companies. The Steering Committee elects one of its members as chairman and appoints a Project Manager. The Project Manager co-ordinates the activities approved by the Steering Committee, including data quality assurance, Det Norske Veritas served as Project Manager during phases | and Tl; SINTEF during phases III - V, and act as current project manager in Phase VII. ®@OREDAOREDA-2002 10 OREDA-2002 EQuipMENT CATEGORIES COVERED IN THE DIFFERENT PHASES: Table | shows the equipment categories that have been included in the four OREDA. handbooks (including this one). Most of the equipment derives from offshore installations, but a few equipment units from onshore E&P have also been included. In phase V more emphasis has been placed on collection of subsea data. i Table 1 Equipment classes covered by the four OREDA Handbooks Phase! ] Phasell | Phaselll | PhasetV | PhaseV | SUM (cae 2| ear X ter Equipment cla edition) | C2830") eciday | (2002 action} System c1963— | (1987-90) |(1990- | ir90a— 35 22) Ne of units [ No-ofunts No, ofunis Gan Tarn 2 = Compressors 1| 5 Combuston engines 38 et] 103 “pumpe as zr] twa} “zae} 52) 1289 “Turbooxpanders 7 3 15 “Eieahi generators 75 * er 8) 20 {Elecre notre 56} zz] = Vessels Es a =a si] ott ste | -Hestereencboiers 3 1 | eatement |-Heatexchangere 9 v0 75 5 a] aw vais 358 oe Se tops |-F86 detecton 23685] 5823 779] 10388 one }equipment + SProcesesersorsconto!| 9740 ser] io oo| aaa Tice TMs a ayslems a2t a7" auipment | Nise. eteysyotoms 1703 703 phase only |= Nise utity eystome 4038 1035 - Driling systems = Control systems = Wallnead & X-mas tree Pipelines Template - Manifold Risers = Running tools + Miss. equipment (phase 0 ? The -92 edition de also contain the data issued in the -84 version, @OREDAOREDA-2002 4 OREDA-2002 SCOPE OF THE OREDA HANDBOOK The OREDA handbook presents high quality reliability data for offshore equipment collected during phase IV and V of the OREDA project. The intention of the handbook is to provide both quantitative and qualitative information as a basis for RAMS analyses. For cach topside equipment unit, the following information is presented: ‘© A drawing illustrating the boundary of the equipment unit, ‘e., a specification of subunits and so-called maintainable items that are part of the equipment unit. «A listing of all failuze modes, classified as critical, degraded or incipient, respectively. © The observed number of failures for each failure mode. ‘The aggregated observed time in service for the equipment unit, classified as calendar time, operational time, and number of demands. ‘© Anestimate of the failure rate for each failure mode with associated unceztainty limits. A tepair time estimate, Le., the number of man-hours required to repair the failure and restore the function. «© A repair time estimate, i¢., the elapsed time in number of hours required repairing the failure and restoring the function. This time is the active repair time, i.c. the time when actual repair work was being done. Supportive information, e.g., number of items and installations. ‘A cross-tabvlation of maintainable item versus failure mode, and of failure descriptor/- cause versus failure mode. For each subsea equipment unit, the following information is presented: © A drawing illustrating the boundary of the equipment unit, i.c., a specification of subunits and components that are part of the equipment unit. © A Listing of all components. The observed number of failures for each component. The aggregated observed time in service for the equipment unit, classified as calendar time. An estimate of the failure rate for each component with associated uncertainty limits. ‘A repair time estimate, i., the elapsed time in number of hours required repairing the failure and restoring the function. This time is the active repair time, i.e. the time when actual repair work wes done. ‘© Supportive information, ¢.g., number of items and installations. ‘A crosstabulation of component versus failure mode, of subunit versus failure mode, of cquipment unit versus failure mode and of failure deseriptor/-cause versus failure mode @ OREDAOREDA- 2002 12 OREDA-2002 Limitations Information released from each participating company has been kept confidential by rendering it anonymous. Only generic data are published. The single event information, which is the basis for the estimates, is (in most cases) gathered from two or more installations, and consequently the figures in the handbook reflect a weighted average of the experience. The OREDA project is so far restricted to failure data collected on hardware components and systems; information about human errors is not included. Nevertheless, contponent failures may have been caused by human errors and, therefore, implicitly, human errors are included in the failure rate estimates. Details and limitations of the methods used are described in the section “ESTIMATION PROCEDURES” on page 23. @OREDAOREDA-2002 13 OREDA-2002 THE OREDA TOPSIDE DATA STRUCTURE GENERAL In order to collect data and analyse them in a consistent manner, a taxonomy description has been developed in the OREDA project. The following gives a summary of that taxonomy to better understand the platform oa which these data have been collected and stored. Note that some of the parameters given in this description are not included in the generic data presented in this handbook, but contained in the source database. MAIN DATA CATEGORIES For each equipment category the database is split into three separate database files: an Imventory part, a Maintenance part, and a Failure part. ‘The Inventory part contains a description of each equipment unit for which data have been collected, e.g.. pump. This description contains technical data {e.g., capacity, size) as well as some operating and environmental data (e.g., operating mode, vibrations). The inventory deseription for each equipment unit is stored in an Inventory record in the database. ‘The Failure part contains the failure events being experienced for an equipment unit (inventory) during the period of surveillance: one record for each failure event. The failure events are always related to one equipment unit (inventory). The Maintenance part contains information about the corrective and the scheduled preventive maintenance program for each equipment unit (e.g,, maintenance action, interval, man-hours). Data on corrective maintenance is related to its preceding failure, while data on preventive maintenance is related to the equipment unit. SYSTEM HIERARCHY The various items are classified into equipment categories termed Equipment classes e.g. pumps, compressors, valves ete, Each individual item within a class is termed an Equipment Unit (e.g. one pump). Each equipment class is further classified according to its design characteristics and type of service (system). Table 2 gives an example for the two equipment classes Pumps and Fire & Gas detectors. Equipment within an equipment class is subdivided in two lower indenture levels, called subumts and maintatnable items (MP). This subdivision is purely hicrarchic and has die following interpretation: Level 1 - Equipment Unit: The highest level used in OREDA and typically includes an equipment unit with one main function, e.g. pump, compressor. @OREDAOREDA- 2002 14 OREDA-2002 Level 2 — Subunit: An equipment unit is subdivided in several subunits, each with one function required for the equipment unit to perform its main function. Typical subunits are €g., cooling, lubrication. The subunits may be redundant, e.g., two indepeadent start units, Level 3 - Maintainable Item (MI): These are subsets of cach subunit and will typically consist of the lowest level units that are due for preventive maintenance. ‘The hierarchy is illustrated in Figure |. Table 2 System Classification (Example) Centrifugal Water fre ighting Reciprocating Sea water nection Rotary Oil handling Gas utilities Gas processing Smoke/combustion Fire detection Heat Flame Hydrocarbon gas Gas detection H2S gas (Example: PU-RO-OH indicate a retary pump used in cl handling) @OREDAOREDA-2002 15. OREDA-2002 Level 1 EQUIPMENT wa 5m Level 2 ‘SUBUNIT 1 2am pore So £ [ aR anne ates o saranas ems 2s waarmee mathe Levels wananane warannais ITEM No. N ITEM No N Figure 1 System Hierarchy Several subunits may be relevant for several equipment categories (e.g., lubrication system, starting system). In these cases the subunits are given the same name and the same set of MIs. This is done in order to standardise the subunits/MIs as much as possible, although some of the Mls in these subunits may not apply for all equipment categories. EquiPMENT BOUNDARIES To compare failure events from different equipment categories, installations, or sources, it is important to have a common definition of which components or parts that are to be included in an inventory. The boundary defines parts associated with the genetic item that are considered ta be essential for its function or that are sold by the manufacturer as part of the item. For example, the power transmission (e.g. gear) is included within the boundary for a pump, while the driver (e.g. elmotor) is not. The boundary is normally sufficiently determined by a boundary diagram as illustrated in Figure 2. Further, a tabular description as shown in Table 3 lists those subunits and Mls that are included within the boundary. The boundaries are established to confine the same items as the corresponding tag numbers of sub-tag numbers used by the participating oil companies. The equipment units correspond to the companies’ main tag level, while the subunits correspond to the sub-tag level. When establishing the equipment boundaries, the following principles have been applied: ‘®OREDAOREDA-2002 16 OREDA-2002 * The connected units are exchided from the equipment unit boundary unless specifically included by the boundary specification. Failures that occur in a connection (e.g. leakj are inchided unless it is known specifically that it has occurred on the connected item outside the boundary. «When a driver and the driven unit use common subunits (¢.g., lubrication), failures of this subunit is as a common rule related to the driven unit. ‘+ Failures on drivers (€.g. gas turbine) and driven units (e.g. compressors) are presented for cach of those equipment classes separately. When e.g. a failure rate for a combination of driver and driven units is needed (e.g. compressors driven by gas turbines) the combined values from those two equipment classes should be used. © Instrumentation is included only where this equipment has specific control and/or monitoring function for the equipment unit and/or is locally mounted (sensors). Instrumentation of a more general use, such as supervisory system (SCADA) is, as a nile, not included. @OREDAOREDA-2002 17 OREDA-2002 EL. powor nm EXHAUST Inlet Outlet j i] power orwer |? STARTING EL] powes ‘SYSTEM (CieselE [7] MISSION PUMP UNIT 6) | 1 | (Gearbox, et.) need : L MONITORING SYSTEM : Hl Remote instr. Power Coolant coe Boundary Figure 2 Boundary Definition, Pumps INVENTORY DATA For each equipment unit there is an inventory description divided into two parts: L. One part common to all equipment categorics (¢.g., manufacturer, model, function, operating time). 2, One part containing equipment category spevifie data (ex., capacity, size, power consumption). @OREDAOREDA-2002 18 OREDA-2002 | Mortering unit | Internal pwr supply FAILURE DATA In OREDA a failure event is defined as a physical failure of equipment. This implies that all events where a work order is issued, and some maintenance action carried out, would be considered as failure in OREDA (see the definitions on page 40). For each failure a description (record) of the failure is given in the database together with the corrective action(s) cartied out to restore the item to normal operating condition, The information is partly based on numeric data, partly on codes selected from a predefined menu, and partly on free text @OREDAOREDA-2002 19 OREDA-2002 THE OREDA SUBSEA DATA STRUCTURE MAIN DATA CATEGORIES The OREDA subsea database consists of three main parts: An Inventory patt, a Failure patt, and a Maintenance part. ‘The Inventory part contains a description of cach Equipment Unit (e.g. an X-mas tree). It contains technical and operational data on all three indenture levels applied: (1) Equipment Unit, (2) Subunit and (3) Component level. ‘The Failure part contains the failure events experienced for one Equipment Unit during the period of surveillance; one failure record for each failure event. Tf no failures are experienced for a specific equipment unit, the corresponding faiture database will be empty. Subsea failures are linked to the lowest level in the equipment hierarchy, the component level. The Maintenance part contains information about the corrective maintenance/ intervention being carried out (e.g. maintenance action, downtime, resources) and is related to a failure event record, ‘SYSTEM HIERARCHY The system hierarchy in OREDA, subsea part, is broken down into four levels starting on top viz: © Field/Installation: This is an identifier for the subsea field and its installation(s). For each field several installations may be included. © Equipment unit: An equipment unit on the highest equipment level used in OREDA. which typically includes a unit with one main function, e.g. X-mas tree, control system, etc. © Subunit: An equipment unit is subdivided in several subunits, each with function(s) required for the equipment unit to perform its main function. Typical subunits are e.g. umbilical, HPU etc. The subunits may be redundant, e.g. two independent HPUs. Component: These are subsets of each subunit and will typically consist of the lowest level items that are being repaired or replaced as a whole (e.g. valve, sensor ete) ‘The hierarchy is illustrated in Figure 3. @OREDAOREDA-2002 20 OREDA-2002 Inventory data Failure data | Maintenance data Bec) ee _— i Cem] Bie wees | | : a sane | z Sakata } sae] ‘ | =— (a Tele nie | [senate ; Figure 3 - System hierarchy The failures and corrective maintenance actions are as shown in Figure 3 linked to the component where they occurred, EuuibMeNnt BOUNDARIES The boundaries of what constitutes a subsea system and the various levels in the inventory need tn be clearly defined to ensure that in-service times and failures ate allocated correctly. A typical equipment level boundary used in OREDA is illustrated in Figure 1, Boundary details for the each equipment class are given in each equipment class chapter respectively. @ ORCDAOREDA- 2002 21 OREDA.2002 EB, | oe eT | =e | pe poets | : ow sve at Son Y= =] onan ' | i sr a 1 om [aca ee, men | chcamnanfs | Legon: —_— ASUESY: Anauba Production Saab Vee AMVAMY. AmubiProdesten Maser Valve AWWPWY) Anmuke’Prodston Wing Vale COV: Ciosover Valve SCSSV: — Suaes Coals Sint Safty Valse Figure 4 - Boundary definition, X-mas tree INVENTORY DATA The inventory data are used to describe an equipment unit and its associated subunits and components, to be able to compare equal with equal and database. Furthermore, the inventory data may represent ex; observed reliability. These data are mainly of static character; ic. they do not ch. retrieve relevant data from the cplanatory variables affecting the ange in course of time. They are also recorded once for each item. Some are, however, of a more dynamic nature and may change during the surveillance period (e.g. no. of demands). ®@OREDAOREDA-2002 22 OREDA- 2002, FAILURE EVENT AND MAINTENANCE DATA For each failure a description of the failure is given together with the maintenance/intervention (corrective action) carried out to restore the item to normal operating conditions. This information is divided in two event revords © Failure; ice. description of the failure event © ~Maintenance/intervention; i.e. description of the maintenance action These records contain a set of attributes describing the failure and maintenance action respectively. The attributes are based on numeric data, codes selected from a predefined menu, and five text description. @OREDAOREDA-2002 23 OREDA-2002 ESTIMATION PROCEDURES The main purpose of the OREDA-2002 handbook is to present average failure rate estimates together with repair time estimates. This section presents a brief description of the statistical methods that are used. FAILURE RATE The failure rare function tells us how likely it is that an item that has survived up to time will fail during the next unit of time, If the item is deteriorating, this likelihood will increase with the age + A man who has reached the age of 95 years will obviously have a higher probability of dying during the next year than a 20 years old man. The failure rate function will therefore usually be a function of the time - or, the age of the item. To give a mathematical definition of the failure rate funetion, we start with the time to failure, 7, of the item, i.e, the time from the item is put into operation until the first failure occurs. It is generally impossible to predict the exact value of the time to failure, and 7 will therefore be a random variable with some distribution. The failure rate function, A{/), may now be detined mathematically as: Atat= Pre< 7s t+ AtlT> ‘The right hand side of this equation denotes “the probability that the item will fail in the time interval (, ¢+ Ad), when the item is still functioning at time ¢” or with other words: “the probability that an item that has reached the age ¢ will fail in the next interval ¢, ¢+ A.” The approximation is sufficiently accurate when AF is the length of a very “short” time interval. The failure rate function is sometimes also called ‘hazard rate’ or ‘force of mortality”. The life of a technical item may generally be split into three different phases: the burz-in (or infant mortality) phase, the usefiz! tife phase, and the wear-out phase, The failure rate function will usually have different shapes in the three phases. As illustrated in Figure 5, the failure rate function may be decreasing in the burn-in phase, clase to constant in the useful life phase, and increasing in the wear-out phase. The curve in Figure 5 is called a “bath-tub” curve because of its characteristic shape, and is often claimed to be a realistic model for ‘mechanical equipment. If we assume that the failure rate function is constant during the useful life phase, this means that the item is not deteriorating during this phase. The deterioration will start when, or if, the item enters the wear-out phase. @OREDAOREDA-2002 24 OREDA-2002 Fallure rate function po Burn-in phase Useful life phase Wear-out phase Time. Figure 5 Bath-Tub Shape of the Failure Rate So-called burn-in problems may be caused by inherent quality problems in the item, or by installation problems. Inherent quality problems may sometimes be removed by carefull quality testing prior to installation. Installation problems have been disregarded in the OREDA data collection, notably for most topside equipment. The bum-in phase is therefore not included in the OREDA database, and we may assume that the data collection is started with the usefull life phase. For subsea equipment data is collected on a whole lifetime basis, i.e. data collection starts when the equipment is installed and ready for its intended service, This means thal the equipment may no necessarily have been Many of the items covered in OREDA are subject to some maintenance or replacement policy. The items will thereby often be replaced or refurbished before they reach the wear-out phase. ‘The main part of the failure events in the OREDA database will therefore come from the ‘useful life phase, where the failure rate is close to constant. All the failure rate estimates presented in this handbook are therefore based on the| [assumption that the failure rate function is constant and independent of time, in which case| Aft) = A. Note «No statistical tests have been performed to verify the assumption of a constant failure rate. * Since data are assumed to come from “bottom” of the bath-tub curve, the failure rate estimates presented therefore represent some kind of minimum over the entire life cycle of the equipment. An important implication of the constant failure rate assumption is that an item is considered to be “as good as new” as long as it is functioning. All failures are purely chance failures and independent of the age of the item. The mean time to failure, MTTF, may be calculated as @OREDAOREDA-2002 2 OREDA-2002 1 MTTF =— a These and related concepts are thoroughly discussed in e.g., Hayland and Rausand (1994) ESTIMATORS AND UNCERTAINTY LIMITS FOR A HOMOGENEOUS SAMPLE When we have failure data from identical items that have been operating under the same operational and environmental conditions, we have a so-called homogeneous sample. The only data we need to estimate the failure rate A in this case, are the observed number of failures, n, and the aggregated time in service, t. The estimator of A is given by: Number of failures # ‘Aggregated time inservice T See e.g. Hoyland and Rausand (1994) for further details. ‘The aggregated time in service, 7, may be measured either as calendar time or operating time, and both these are presented in the data tables in Part IL Note that this approach is valid only in the following situations; «Failure times for a specified number of items, with the same failure rate A, are available. ‘© Data (several failures) is available for one item for a period of time, and the failure rate is constant during this period. © A combination of the two above situations, ie., there are several items where each item might have several failures. This is the typical situation for the OREDA data. In the data tables in Part 1] of the handbook, estimates are given for each failure mode. Uncertainty intervals for the failure rate “The uncertainty of the estimate 4 may be presented as a 90% confidence interval. This is an interval (A,,Ay), such that the “truc value” of A fulfils: Pr(Ag SA < Av} = 90% With n failures during an aggregated time in service +, this 90% confidence interval is given by: : ) ‘20882095 Zahedi au @ OREDAOREDA-2002 26 OREDA-2002 where zee, and zoocy denote the upper 95% and 5% percentiles, respectively, of the 7° distribution with v degrees of freedom, see Table 4, page 31. Example Assume that n = 6 failures have been observed during an aggregated time in service = 10000 hours. The failure rate estimate is then given by: A = ni t= 6-10" failures per hour and a 90% confidence interval is given by: L L L 1 4 . Fy 2008 3 o0sz0ry | = oats znenne] = 2.6104, 11,810 (spe nn ] (saan 20000“ ] ¢ ) ‘The estimate and the confidence interval are illustrated in Figure 6. — > SOT 10 11 12 (fallures per 16 nous) Figure 6 Estimate and 90% Confidence Interval for the Example. Note The given interval is a confidence interval for the failure rate for the items we have data for. There is no guarantee that items installed in the future will have a failure rate within this ‘interval. MULTI-SAMPLE PROBLEMS In many cases we do not have a homogeneous sample of data. The aggregated data for an item may come from different installations with different operational and environmental conditions, or we may wish to present an “average” failure rate estimate for slightly different items. In these situations we may decide to merge several mote or less homogeneous samples, into what we call a multi-sample. The various samples may have different failure rates, and different amounts of data - and thereby different confidence intervals. This is illustrated in Figure 7. @OREDAOREDA-2002 ar OREDA- 2002 Faure rate 7 2 8 4 8 BF BB HDA Ag HatBE AT I Fore) Figure 7 Multi-Sample Problem ‘To merge all the samples, and estimate the “average” failure rate as the total number of {failures divided by the aggregated time in service will not always give an adequate result. The ‘confidence’ interval will especially be unrealistically short, as illustrated in Figure 7. We therefore need a more advanced estimation procedure to take care of the multi-sample problem. Below, the so-called OREDA-estimator of the “average” failure rate in a multi-sample situation is presented together with a 90% uncertainty interval. Spjotvoll (1985) gives a rationale for the estimation procedure. The OREDA-estimator is based on the following assumptions: © We have & different samples. A sample may ¢.g., correspond to a platform, and we may have data from similar items used on k different platforms. © Insample no, i we have observed 1; failures during a total time in service 1, for i-12,...,k © Sample no. i has a constant failure rate Ay, for i =1,2..... & © Due to different operational and environmental conditions, the failure rate A; may vary between the samples, The variation of the failure rate between samples may be modelled by assuming that the failure rate is a random variable with some distribution given by a probability density function nth). ‘The mean, or “average” failure rate is then: fe en(A) dd. and the variance is: ot =f(2-0,) (Aad. + © OREDAOREDA-2002 28 OREDA-2002 ‘To calculate the multi-sample OREDA-estimator, the following procedure is used: 1. Calculate an initial estimate 4, of the mean (“average”) failure rate 8, by pooling the data : Sn Total-no. of failures _ ‘Total time in service a 2. Calculate: b = Ye a ye i S: ya yaetak yt as, ti at 3. Calculate an estimate for 0°, a measure of the variation between samples, by: V- k-0)6; x 8 ‘Si-Sz v when greater than 0, else 0. 4. Calculate the final estimate 6° of the mean (“average”) failure rate @ by: In the data tables in Part Il of the handbook @" corresponds to the mean (column 4), and ‘SD corresponds to the standard deviation (column 6) The dower and upper “uncertainty” values are given by: ope J 7A) di=90% Kemer @OREDAOREDA-2002 29 OREDA-2002 Since the distribution m(A) is not known in advance, the following pragmatic approach is used: 6. m(X) is assumed to be the probability density function of a Gamma distribution with parameters and 8. 7, The parameters cand f are estimated by: i Uppers avesat where Zossy and Zoos denote the upper 95% and 5% percentiles, respectively, of the {C-distribution with v degrees of freedom, see Table 4, page 31. In situations where v is not an integer, an interpolation in the 77-distribution is performed. Note I More detailed analysis of the OREDA data (see Vatn 1993) has indicated that there may be a large variation between installations. The multi-sample OREDA estimator should therefore as a rule be used instead of the 1/7 estimator which is based on @ homogeneous sample. The variation between the samples (installations) is measured by the standard deviation SD, Note 2 In the OREDA-84 and OREDA-92 handbooks, a slightly different approach was taken. The mean value was estimated with the same procedure as in this handbook, but the Jower and upper values were given a slightly different interpretation. Note 3 In the case of k= 1, the procedure cannot be used. In this case the n/t estimate is given for the mean, and the /ower and upper values should be interpreted as a traditional 90% confidence interval. Note 4 If no failures are observed for an item, the following approach is used to obtain lower, mean and upper values for “All failure modes": 1, Let Ay denote the failure rate estimate (“mean”) one level up in the taxonomy hierarchy. @OREDAOREDA.2002 30 OREDA-2002 2. Let 7 denote the total time in service (operational or calendar) for the item of interest 3. Let a= 12 B=—-+t 4, An estimate for the failure rate is now aet 8 5. The standard deviation is given by w= fe 6. A.90% uncertainty interval is given by ose te: = ( 2.002 1.9. ppree age Fla ESTIMATION OF DEMAND PROBABILITIES. If information about “number of demands” is given (see Section “Data table, Reliability Data”, page 32) it is possible to estimate the demand probability. The demand probability is always related to one specific failure made, for example a critical fail to start. The demand failure probability is estimated by: . om a where n is the number of failures with the appropriate failure mode, and d is the number of demands, Note that in the data table presentations the demand probabilities may apparently look different. The reason for this is that in some cases there are registered “demand failures", but the number of demands is not recorded for one or more inventories. For these inventories, the demand failures are not added to the total number of demand failures for that data table. @OREDAOREDA-2002 31 PERCENTAGE POINTS OF THE CHFSQUARE DISTRIBUTION Table 4 Percentage Points of the Chi-square (x2) Distribution OREDA-2002 wo Sewrsusune i 12 13 14 15 16 17 18 19 20 25 26 27 28 29 30 60 10 80. 100, 0.00 0.10 0.35 0.71 1.15 1.64 217 2.73 3.33 3.94 4.57 3.23 5.89 6.57 7.26 7.96 8.67 9.39 10.12, 10.85 14.61 15.38 16.15 16.93 17.71 18.49 26.51 34.76 ABAD 51.74 60.39 69.13 77.93 3.84 5.99 781 9.49 11.07 12.59 14.07 15.51 16.92 18.31 19.68 21.03 22.36 23.68 25.00 26.30 27.59 28.87 30.14 3141 37.65 38.89 40.11 A134 42.56 43.77 55.76 67.50 79.08 90.53 101.88 113.14 124.34 @OREDA 5.02 7.38 935 114 12.38 14.45 16.01 17.53 19.02 20.48 21.92 23.34 24.74 26.12 27.49 28.85 30.19 31.53 32.85 34.17 40.65 41.92 43.19 44.46 45.72 46.98 59.34 71.42 83.30 95.02 106.63 118.14 129.56 9.21 11.34 13.28 15.09 16.81 18.48 20.09 21.67 23.21 24,72 26.22 27.69 29.14 30.58 32,00 33.41 34.81 36.19 37.57 44.31 45.64 46.96 48.28 49.59 50.89 $3.69 76.15 88.38 100.42 112.33 124.12 135.81 116.32, 128.30 140.17OREDA-2002 32 OREDA-2002 TOPSIDE DATA TABLE FORMATS DATA TABLE, RELIABILITY DATA Each data table contains an identification of the item and the estimated reliability parameters. ‘The figures provided should be interpreted on the basis of the assumptions specified in the boundary definition for each equipment category and the estimation method applied. The format of the data table is shown in Figure & Taxonomy ne Tem Population | Installations Aggregated tne n service 10 hours) oof domands Calender time” ‘Operational time * Faure mode Woo! Faire rate (per 10" hours) ‘ative [__ Repair Granhours] faites | Lower | Mean | Upper | SO [mit | tops [Min | Mean | Max [Comments Figure 8 Format of the Reliability Data Tables The various entries of the data table are explained in the following: Taxonomy number and Item ‘The taxonomy number is a numerical identification of the item. The description of the item is given in a hierarchical structure. Only data from items of this generic category of compon- ents/equipment are input (o the estimates presented in the quantitative part of the data table. Poputation Total number of items forming the basis for the estimates. Installations Total number of installations (platforms) covered by the data surveillance for the item in question. ‘®@OREDAOREDA-2002 ES} OREDA-2002 Aggregated time in service Two types of time scales are presented as the basis for the failure rate estimates; calendar time and operational time, The aggregated time in service for the total population is given for both time scales. Note that while the calendar time is given with high certainty, the operational time has in many cases to be based on estimates (by the data collector). Number of demands The accumulated number of demands/cycles for the total population is given when available. In several cases these numbers are based on estimates and not accurate measurements. Failure mode This column contains a brief description of the manner in which the failure occurred, when such information is available. Number of failures: The total number of failure events is presented for each failure mode, The accumulated number of failures is presented as “All modes”, Failure rate The failure rate columns present estimates of the failure rate for each failure mode. Results are given both under the “multi-sample” assumption, and under the assumption of homogeneous data sets. In the multi-sample situation the failure rate is assumed to vary between installations (platforms), and each platform represents one sample. The following entries are included: Mean ‘An estimate of the “average” failure rate with respect to the specified failure mode, obtained by using the OREDA estimator. (Lower, Upper) A 90% uncertainty interval for the failure rate. sD A standard deviation indicating the variation between the multiple samples. ad ‘The total number of failures divided by the total time in service, ie., the estimate of the failure rate we would use for a homogeneous sample. ‘All the entries are measured per 10® hours and refer either to calendar time (marked *) or operational time (marked 4). @OREDAOREDA-2002 34 OREDA-2002 Active repair time (hours) This column contains the average calendar time (hours) required to repair and retum the item. toa state where it is ready to resume its functions. Active repair time is the time when actual repair work is being done. It does not include time to shut down the unit, issue work order, wait for spare parts, start-up after repair etc. The active repair time is therefore normally shorter than the downsime where some of the activities indicated above may be included. Note: During the data collection exercises it has been very difficult to obtain data regarding active repair times. In the OREDA database there is a good coverage of “restoration man-hours” data, whereas the data for “active repair time” is rather sparse. It should also be noted that active repair hours are highly influenced by how maintenance is organised on the platform, The figures for active repair times should therefore only be used as an indication of what the actual active repair times would be. It is highly recommended to use some kind of expert judgement in addition to the values given in the handbook. Repair (manhours) ‘The repair columns present three values of the repair time (man-hours). The mean value is the average number of man-hours recorded to repair the failure and restore the function, The min and max values are the lowest and highest number of man-hours recorded for the repair of the item. Comments ‘When available the on on-demand failure probability is given in the Comment field. DATA TABLE, MAINTAINABLE ITEM VERSUS FAILURE MODE The reliability data presented in the data table in Figure 8 (page 32) does not give information on which part of the equipment has failed. In the Maintainable Item versus Failure Mode listing the relative contribution from each maintainable item to the total failure rate may be obtained. The figures in the table represent percentages of occurrence for each combination of failure mode and maintainable item. The row sum represents the total percentage of failures that are related to the actual maintainable item. Note that several maintainable items might be assigned to each failure record. In such situations, the “score” for the actual maintainable ite/failure mode combination is set to l/n, where m is the number of maintainable items listed for that failure record. The column sum represents the contribution for each failure mode in percentages. This information is valuable input to an FMEA/FMECA analysis. The FMEA/FMECA analysis is further a major part of a reliability centred maintenance (RCM) analysis. As the RCM methodology focuses on failure causes, it is also important to have information regarding failure causes as discussed in the next section. Note that several maintainable items might be assigned to each failure record. @OREDAOREDA-2002 36, OREDA-2002 DATA TABLE, FAILURE DESCRIPTOR VERSUS FAILURE MODE In the Failure Descriptor versus Failure Mode listing the relative contribution from each failure descriptor (cause) to the to the total failure rate may be obtained. The figures in the table represent percentages of occurrence for each combination of failure descriptor and failure mode. The row sum represents the total percentage of failures that are related to the actual failure descriptor. The column sum represents the contribution for each failure mode in percentages. As mentioned above, the information about failure causes is essential in an RCM analysis. For example scheduled replacement of units is only applicable if one or more failure causes may be related to ageing, wear, corrosion etc. © OREDAOREDA-2002 3 OREDA-2002 SUBSEA DATA TABLE FORMATS DATA TABLE, RELIABILITY DATA Each data table contains an identification of the item and the estimated reliability parameters, The figures provided should be interpreted on the basis of the assumptions specified in the boundary definition for each equipment category and the estimation method applied. The format of the data table is shown in Figure 8. Taxonomy na Icam Population | Installations ‘Aggregated time inservice (1 hours) lure dat Calendar tw Component | Noof | | Severity class Failure rate (per 10" hours), units #icloli | u [tower [Mean | Uppe: | SD [Subunit no. 1 Compan na 7 [Componene na.2 [Subunit no. 2 Component ne. Equipment ive [Commams. Foccomponens wih notalues, nis set 205 besed.an a non normative pre, Mes fan forthe commen compares wed te esnsit Figure 9 Format of the Reliability Data Tables The various entries of the data table are explained in the following: Taxonomy number and Item ‘The taxonomy number is a numerical identification of the item. The description of the item is given in a hierarchical structure. Only data from items of this generic category of compon- ents/equipment are input to the estimates presented in the quantitative part of the data table. Population Total number of items forming the basis for the estimates. Installations Total number of installations (platforms) covered by the data surveillance for the item in question. © OREDA