Springer Series in Reliability Engineering Jan Erik Vinnem Offshore Risk Assessment Principles,Modelling and Applications of QRA Studies 2nd Edition
Series EditorProfessorHoang PhamDepartment ofIndustrial EngineeringRutgersThe State University of New Jersey96 Frelinghuysen RoadPiscataway, NJ 08854-8018USAOthertitles inthis seriesThe Universal GeneratingFunction in Reliability Analysis and OptimizationGregory LevitinWarrantyManagementandProductManufactureD.N.P Murthy and Wallace R. BlischkeMaintenanceTheoryof ReliabilityToshio NakagawaSystem SoftwareReliabilityHoang PhamReliabilityandOptimal MaintenanceHongzhouWangand HoangPhamAppliedReliabilityandQualityB.S. DhillonShock and Damage Models in Reliability TheoryToshioNakagawaRisk ManagementTerje Aven and Jan Erik VinnemSatisfyingSafetyGoalsbyProbabilisticRiskAssessmentHiromitsu Kumamoto
Series Editor Professor Hoang Pham Department of Industrial Engineering Rutgers The State University of New Jersey 96 Frelinghuysen Road Piscataway, NJ 08854-8018 USA Other titles in this series The Universal Generating Function in Reliability Analysis and Optimization Gregory Levitin Warranty Management and Product Manufacture D.N.P Murthy and Wallace R. Blischke Maintenance Theory of Reliability Toshio Nakagawa System Software Reliability Hoang Pham Reliability and Optimal Maintenance Hongzhou Wang and Hoang Pham Applied Reliability and Quality B.S. Dhillon Shock and Damage Models in Reliability Theory Toshio Nakagawa Risk Management Terje Aven and Jan Erik Vinnem Satisfying Safety Goals by Probabilistic Risk Assessment Hiromitsu Kumamoto
PrefacetolstEditionFrom amodest start in Norway as a researchtool in the late1970s,Quantified RiskAssessment (QRA)for offshore installations has become a key issue in themanagement of Safety,Health and Environment in the oil and gas industriesthroughout the entire North Sea. While the initiatives in the early stages often camefromtheauthorities,theuseisnowmainlydrivenbytheindustryitself.TheQRAis seen as a vehicle to gain extended flexibility with respect to achievement of anacceptable safety standardin offshoreoperations.Themodelsmaybeweak insomeareas and theknowledge is sometimes limited,butstudiesareneverthelessused effectively in the search for concept improvement and optimisation of designandoperation.This book results from working with offshore QRAs for more than 20 years.The author has, during this period, had the opportunity to practice and evaluate theuse of such studies from different perspectives,the consultancy's, the operatingcompany's, the researcher's and the educator's point of view.The author has for several years taught a course in risk analysis of marinestructures at theFaculty of Marine Technology,NTNU, Trondheim, Norway.ThestartingpointforthemanuscripwastheneedtoupdatethelecturenotesItishopedthatthisbookinthefuturealsomaybeausefulreferencesourceforawider audience.Therehas beeyearsarapidexpansionoftheuseofforsomerisk assessments for the offshore oil and gas activities.It is expected that theexpansion is going to continue for some time, as the offshore petroleum industryexpands into new regions and meets newchallenges in old regions.The oil pricereached its lowest level for manyyears,during the first quarter of1999.Onemightbetemptedtothinkthatthe economic climatemayprohibitfurther attention to risk assessment and safety improvement.The oppositeisprobably more correct. As a friend in Statoil expressed not so long ago:"Wheneverthemarginsaregettingtighter,theneedforrisk assessments increases,asnewandmore optimised solutions are sought, each needing an assessment ofrisk'InNorway,thebeginningof1999 is alsothetimewhen theNorwegianPetrole-um Directorate is preparing a major revision of the regulations for offshoreinstallations and operations,anticipated to comeintoeffect in2001.Ithas obvious-lynotbeenpossibletocapturethefinalrequirementsofthenewregulations,butan
Preface to 1st Edition From a modest start in Norway as a research tool in the late 1970s, Quantified Risk Assessment (QRA) for offshore installations has become a key issue in the management of Safety, Health and Environment in the oil and gas industries throughout the entire North Sea. While the initiatives in the early stages often came from the authorities, the use is now mainly driven by the industry itself. The QRA is seen as a vehicle to gain extended flexibility with respect to achievement of an acceptable safety standard in offshore operations. The models may be weak in some areas and the knowledge is sometimes limited, but studies are nevertheless used effectively in the search for concept improvement and optimisation of design and operation. This book results from working with offshore QRAs for more than 20 years. The author has, during this period, had the opportunity to practice and evaluate the use of such studies from different perspectives; the consultancy’s, the operating company’s, the researcher’s and the educator’s point of view. The author has for several years taught a course in risk analysis of marine structures at the Faculty of Marine Technology, NTNU, Trondheim, Norway. The starting point for the manuscript was the need to update the lecture notes. It is hoped that this book in the future also may be a useful reference source for a wider audience. There has been for some years a rapid expansion of the use of risk assessments for the offshore oil and gas activities. It is expected that the expansion is going to continue for some time, as the offshore petroleum industry expands into new regions and meets new challenges in old regions. The oil price reached its lowest level for many years, during the first quarter of 1999. One might be tempted to think that the economic climate may prohibit further attention to risk assessment and safety improvement. The opposite is probably more correct. As a friend in Statoil expressed not so long ago: ‘Whenever the margins are getting tighter, the need for risk assessments increases, as new and more optimised solutions are sought, each needing an assessment of risk’. In Norway, the beginning of 1999 is also the time when the Norwegian Petroleum Directorate is preparing a major revision of the regulations for offshore installations and operations, anticipated to come into effect in 2001. It has obviously not been possible to capture the final requirements of the new regulations, but an
viPrefaceattempthas been madeto capturethenewtrends intheregulations,totheextentthey areknown.There have over the last 10-15 years been published a few textbooks on riskassessment,most of them are devoted to relatively generic topics.Some are alsofocused ontheriskmanagementaspects,ingeneral and withoffshoreapplicability.None areknown to address the needs and topics of the use of QRA studies by theoffshore industry in particular.The present work is trying to bridge this gap.The use of QRA studies is somewhat special in Northern Europe,and parti-cularly in Norway.The use of these techniques is dominated by offshore applica-tions, with the main emphasis on quantification of risk to personnel.Furthermore,therisk to personnel is virtually never concerned with exposure of the public tohazards.Thus,thestudiesarerarelychallengedfromamethodologypointofviewMostpeoplewill probably seethis as anadvantage,but it alsohas somedraw-backs.Suchchallengesmaynamelyalsoleadto improvements inthemethodology.Itmaynotbe quite coincidental that the interest in modelling improvementand development sometimes has been rather low between the risk analysts workingwith North Sea applications.This book attempts to describe the state-of-the-art with respect to modelling inQRA studies for offshore installations andoperations.It also identifies some of theweaknessesand areaswherefurtherdevelopment shouldbemade.Ihopethatfurther improvement may be inspired through these descriptions.0.1About the ContentsAQuantified RiskAssessmentof an offshore installationhas thefollowingmainsteps:1.Hazard identification2.Cause and probability analysis3.Accidental scenarios analysis4.Consequence,damageandimpairmentanalysis5.Escape,evacuationandrescueanalysis6.Fatalityrisk assessment7.AnalysisofriskreducingmeasuresThis book is structured in much the same way.There is at least one chapter(sometimesmore)devotedtoeach of thedifferent steps,inmainlythe sameorderas mentioned above. Quite a few additional chapters are included in the text, onrisk analysis methodology,analytical approaches for escalation, escape,evacuationandrescueanalysisofsafetyandemergencysystems,aswellasriskcontrolIt isimportant tolearnfrompast experience,particularly from previousaccidents.A dozen major accidents, mainly from the North Sea, arediscussed attheendof Chapter4,in ordertodemonstratewhatproblemsthathavebeenexperienced in the past
vi Preface attempt has been made to capture the new trends in the regulations, to the extent they are known. There have over the last 10-15 years been published a few textbooks on risk assessment, most of them are devoted to relatively generic topics. Some are also focused on the risk management aspects, in general and with offshore applicability. None are known to address the needs and topics of the use of QRA studies by the offshore industry in particular. The present work is trying to bridge this gap. The use of QRA studies is somewhat special in Northern Europe, and particularly in Norway. The use of these techniques is dominated by offshore applications, with the main emphasis on quantification of risk to personnel. Furthermore, the risk to personnel is virtually never concerned with exposure of the public to hazards. Thus, the studies are rarely challenged from a methodology point of view. Most people will probably see this as an advantage, but it also has some drawbacks. Such challenges may namely also lead to improvements in the methodology. It may not be quite coincidental that the interest in modelling improvement and development sometimes has been rather low between the risk analysts working with North Sea applications. This book attempts to describe the state-of-the-art with respect to modelling in QRA studies for offshore installations and operations. It also identifies some of the weaknesses and areas where further development should be made. I hope that further improvement may be inspired through these descriptions. 0.1 About the Contents A Quantified Risk Assessment of an offshore installation has the following main steps: 1. Hazard identification 2. Cause and probability analysis 3. Accidental scenarios analysis 4. Consequence, damage and impairment analysis 5. Escape, evacuation and rescue analysis 6. Fatality risk assessment 7. Analysis of risk reducing measures This book is structured in much the same way. There is at least one chapter (sometimes more) devoted to each of the different steps, in mainly the same order as mentioned above. Quite a few additional chapters are included in the text, on risk analysis methodology, analytical approaches for escalation, escape, evacuation and rescue analysis of safety and emergency systems, as well as risk control. It is important to learn from past experience, particularly from previous accidents. A dozen major accidents, mainly from the North Sea, are discussed at the end of Chapter 4, in order to demonstrate what problems that have been experienced in the past
PrefaceviiThemainhazardstooffshorestructures arefireexplosion,collisionandfallingobjects.Thesehazards andtheanalysis of them arediscussed in separatechapters.Riskmitigationand control arediscussed intwo chapters,followedbyanoutlineof an alternative approach to risk modelling,speciallyfocused on risk relatingtoshort duration activities. Applications to shipping are finally discussed, mainlyrelatingtoproductionandstoragetankers,butalsowithaviewtoapplicationstoshipping ingeneral.0.2AcknowledgementsParts of the material used in developing these chapters were initiallyprepared for acourse conductedforPETRAD(ProgramforPetroleum ManagementandAdmini-stration),Stavanger,Norway.ManythankstoPETRADforallowingthematerialtobeusedinothercontexts.Some of the studies that have formed the main input to the statistical overviewsectionswerefinancedby Statoil, NorskHydro,SagaPetroleum,EifPetroleumNorge and the Norwegian Petroleum Directorate.The author is grateful that thesecompanies have allowed these studies tobe made publically available.Direct financial support was received from Faculty of Marine Technology,NTNU,this is gratefullyacknowledged.My part time position as Professor atFaculty of Marine Technology,NTNU, has also given the opportunity to devotetime to prepare lecture notes and illustrations over several years.The consultancywork in Preventor Ashas nevertheless financed the majority of the work, inclu-ding the external services.Thanks arealso expressedtoKluwer AcademicPublishers,Dordrecht,TheNetherlands,for agreeing to publish this book,and for providing inspiring andvaluableadvicethroughouttheprocess.Appendix A presents an overview of some of the important softwaretools thatarecommonlyused.Thanksto all theconsultancies and software supplierswhohaveprovidedtheinformationrequiredforthisappendix.AppendixBisadirectcopyof the normativetext in theNORSOKGuidelineforRisk and EmergencyPreparedness analysis,reproduced with kind permission from the NORSOKsecretariat.Some of the consultancies have kindly given permission to use some of theirmaterial, theirkind assistance is hereby beinggratefully acknowledged.DNV shallbethankedforallowingtheirdatabaseWorldwideOffshoreAccidentDatabank(WOAD)to beused free of charge,as input to the statistics in the book.TheFireResearch Laboratory at SINTEF has given kind permission to use illustrationsfromtheir fireon sea research,and Scandpowerhas grantedpermission touseanillustration of the risk assessment methodology.SafetecNordic has givenkindpermissiontouseresults andillustrationsfromtheirsoftwareCollide.Iam particularlyindebted to several personswhohave offered veryvaluablehelp in turning this into a final manuscript.My colleague Dr.David Bayly,CrandonConsultants,has reviewed therawmanuscriptand contributed with manyvaluable comments of botha technical and linguistic nature.The importance ofprovidingclearand concisetextcanneverbeoverestimated,theeffortsmadein
Preface vii The main hazards to offshore structures are fire, explosion, collision and falling objects. These hazards and the analysis of them are discussed in separate chapters. Risk mitigation and control are discussed in two chapters, followed by an outline of an alternative approach to risk modelling, specially focused on risk relating to short duration activities. Applications to shipping are finally discussed, mainly relating to production and storage tankers, but also with a view to applications to shipping in general. 0.2 Acknowledgements Parts of the material used in developing these chapters were initially prepared for a course conducted for PETRAD (Program for Petroleum Management and Administration), Stavanger, Norway. Many thanks to PETRAD for allowing the material to be used in other contexts. Some of the studies that have formed the main input to the statistical overview sections were financed by Statoil, Norsk Hydro, Saga Petroleum, Elf Petroleum Norge and the Norwegian Petroleum Directorate. The author is grateful that these companies have allowed these studies to be made publically available. Direct financial support was received from Faculty of Marine Technology, NTNU, this is gratefully acknowledged. My part time position as Professor at Faculty of Marine Technology, NTNU, has also given the opportunity to devote time to prepare lecture notes and illustrations over several years. The consultancy work in Preventor AS has nevertheless financed the majority of the work, including the external services. Thanks are also expressed to Kluwer Academic Publishers, Dordrecht, The Netherlands, for agreeing to publish this book, and for providing inspiring and valuable advice throughout the process. Appendix A presents an overview of some of the important software tools that are commonly used. Thanks to all the consultancies and software suppliers who have provided the information required for this appendix. Appendix B is a direct copy of the normative text in the NORSOK Guideline for Risk and Emergency Preparedness analysis, reproduced with kind permission from the NORSOK secretariat. Some of the consultancies have kindly given permission to use some of their material, their kind assistance is hereby being gratefully acknowledged. DNV shall be thanked for allowing their database Worldwide Offshore Accident Databank (WOAD) to be used free of charge, as input to the statistics in the book. The Fire Research Laboratory at SINTEF has given kind permission to use illustrations from their fire on sea research, and Scandpower has granted permission to use an illustration of the risk assessment methodology. Safetec Nordic has given kind permission to use results and illustrations from their software Collide. I am particularly indebted to several persons who have offered very valuable help in turning this into a final manuscript. My colleague Dr. David Bayly, Crandon Consultants, has reviewed the raw manuscript and contributed with many valuable comments of both a technical and linguistic nature. The importance of providing clear and concise text can never be overestimated, the efforts made in