COST Action C25 Sustainability of Constructions Integrated Approach to Life-time Structural Engineering Volume 2 Summary Report of the Cooperative Activities Integrated Approach to Life-time Structural Engineering Editors: Lus Bragana, Heli Koukkari, Raffaele Landolfo, Viorel Ungureanu, Erkki Vesikari, Oliver Hechler COST is supported by the EU RTD Framework Programme and ESF provides the COST Office through an EC contract COST - the acronym for European COoperation in Science and Technology - is the oldest and widestEuropeanintergovernmentalnetworkforcooperationinresearch.Establishedbythe MinisterialConferenceinNovember1971,COSTispresentlyusedbythescientific communitiesof36Europeancountriestocooperateincommonresearchprojectssupportedby national funds. The funds provided by COST - less than 1% of the total value of the projects - support the COST cooperationnetworks,COSTActions,throughwhich,withonlyaround20millionperyear, morethan30.000Europeanscientistsareinvolvedinresearchhavingatotalvaluewhich exceeds2billionperyear.ThisisthefinancialworthoftheEuropeanaddedvaluewhich COST achieves. Abottomupapproach(theinitiativeoflaunchingaCOSTActioncomesfromtheEuropean scientiststhemselves),lacarteparticipation(onlycountriesinterestedintheAction participate),equalityofaccess(participationisopenalsotothescientificcommunitiesof countriesnotbelongingtotheEuropeanUnion)andflexiblestructure(easyimplementation and light management of the research initiatives) are the main characteristics of COST. AsprecursorofadvancedmultidisciplinaryresearchCOSThasaveryimportantroleforthe realisation of the European Research Area (ERA) anticipating and complementing the activities oftheFrameworkProgrammes,constitutingabridgetowardsthescientificcommunitiesof emergingcountries,increasingthemobilityofresearchersacrossEuropeandfosteringthe establishment of Networks of Excellence in many key scientific domains such as: Biomedicine andMolecularBiosciences;FoodandAgriculture;Forests,theirProductsandServices; Materials,PhysicsandNanosciences;ChemistryandMolecularSciencesandTechnologies; EarthSystemScienceandEnvironmentalManagement;InformationandCommunication Technologies; Transport and Urban Development; Individuals, Societies, Cultures and Health. It covers basic and more applied research and also addresses issues of pre-normative nature or of societal importance. Summary Report of the Cooperative Activities of the COST Action C25 Sustainability of Constructions - Integrated Approach to Life-time Structural Engineering Volume 2 Sustainability of Constructions - Integrated Approach to Life-time Structural Engineering Editors:LusBragana,HeliKoukkari,RaffaeleLandolfo,ViorelUngureanu,ErkkiVesikari, Oliver Hechler Cover Design: Bogdan Oprescu Cover illustration is based on Happy Street, thArchitect: John Kormeling; Structural DesignerRijk Blok (C25 Member and Chair of WG1) The production of this publication was supported by COST: www.cost.eu ISBN: 2011 The authors and the Editors All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without prior written permission from the publisher. LEGAL NOTICE TheEditors,theAuthorsandthepublisherarenotresponsiblefortheusewhichmightbemadeofthefollowing information. 978-99957-816-2-0February 2011 e Dutch Pavilion for World Expo 2010 in Shanghai; Printed by Gutenberg Press Ltd, Malta. Published by: Department of Civil & Structural Engineering, Faculty for the Built Environment, University of Malta, Malta. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction iContents Foreword.............................................................................................. ......................................... v L. Bragana Preface......................................................................................................................................... vii R. Landolfo Chapter 1.State-of-the-ArtReportonServiceLifePredictionandDesign Methodologies Introduction............................................................................................................................. 1 R. Landolfo & E. Vesikari Part 1. Current Methodologies for Service Life Prediction Service Life Prediction Methods Referred to by Eurocode 0 ................................................. 2 E. Vesikari & R. Landolfo Standardization of Service Life Methodologies.................................................................... 10 W. Trinius & C. Sjstrm Part 2. Service Life Design Methods of Constructions Service Life Design Methods for Civil Engineering Structures ........................................... 17 R. Landolfo & E. Vesikari Risk Based Approach to Service Life Assessment of Building Structures........................... 28 Sz. Wolinski Chapter 2. State-of-the-ArtReportonDeterministicandProbabilisticDegradation Models Introduction........................................................................................................................... 39 E. Vesikari & R. Landolfo Part 1. Degradation Models for Service Life Design Degradation Models of Concrete Structures ......................................................................... 40 E. Vesikari, Ch. Giarma & J. Bleiziffer Modelling of Corrosion Damage for Metal Structures ......................................................... 51 R. Landolfo, L. Cascini & F. Portioli Fatigue of Steel and Composite Bridges............................................................................... 61 U. Kuhlmann, H.-P. Gnther, J. Raichle, & M. Euler Degradation modes and models for masonry structures ....................................................... 69 Z. Lj. Bozinovski, J. Lahdensivu & E. Vesikari Durability and service life of wood structures and components -State of the art ................. 81 H.A. Viitanen, T. Toratti, R. Peuhkuri, T. Ojanen & L. MakkonenPart 2. Application of Degradation Models to Service Life Design Service life design of reinforced concrete structures - Alternative probabilistic approaches............................................................................................................................. 94 Sz. Wolinski Service Life Design of Metal Structures Based on Lifetime Safety Factor Method........... 106 L. Cascini, F. Portioli, R. Landolfo iiChapter 3. Survey and condition assessment of structures Introduction......................................................................................................................... 117 G. Hauf & O. Hechler Part 1. Survey and condition assessment of structures by monitoring Monitoring an introduction.............................................................................................. 120 G. Hauf & O. Hechler Part 2. Damage monitoring and control of materials and structures Monitoring of structural behaviour of buildings................................................................. 130 G. Fabbrocino & C. Rainieri Wireless sensor networks: A new tool for structural monitoring........................................ 142 G. Feltrin, R. Bischoff, J. Meyer & O. Saukh Structural Damage and Vulnerability Assessment for Service Life Estimationthrough MEDEA tool.......................................................................................................... 154 G. Zuccaro & M. F. Leone Part 3. Condition assessment Survey and Condition assessment....................................................................................... 169 B. Khn & O. HechlerConcrete deterioration: Specification and assessment ........................................................ 179 A. Taushanov & Ch. Giarma Condition assessment and service-life prediction of reinforced concrete structures: description and application examples of a method developed for aged and new buildings ............................................................................................................... 192 T. Teruzzi Part 4. Examples MEDEA: a multimedia and didactic handbook for structural damage andvulnerability assessment LAquila Case Study................................................................ 201 G. Zuccaro, F. Cacace & M. Rauci Conclusions......................................................................................................................... 214 G. Hauf & O. Hechler Chapter 4. Maintenance, repair and rehabilitation techniques and planning Introduction......................................................................................................................... 221 O. Hechler Part 1. General Management systems for maintenance, repair and rehabilitation ....................................... 223 E. Vesikari, A. Borrmann & K. LukasPart 2. Buildings Maintenance, repair and rehabilitation of buildings ........................................................... 235 V. Ungureanu, A. Dogariu, D. Dubina, A. Ciutina, R. Landolfo, F.M. Mazzolani & F. Portioli Swiss codes for existing structures Principles and challenges......................................... 282 E. Brhwiler, T. Vogel, T. Lang & P. LchingerPart 3. Bridges Bridge Maintenance, Repair and Rehabilitation in a Life-Cycle Context .......................... 288 D.M. Frangopol & N.M. OkashaiiiStrengthening of civil structures ......................................................................................... 296 J. Andersson, A. Geler, J. Naumes & O. Hechler Conclusion .......................................................................................................................... 310 O. Hechler Chapter 5. Demolition and deconstruction Introduction......................................................................................................................... 315 F. Portioli & O. HechlerPart 1. GeneralSustainable refurbishment, deconstruction and demolition: Applications of MCDMmethods ............................................................................................................................... 317 F. Portioli, L. Cascini, R. Landolfo & V. UngureanuAssessment of the structural materials at high stain-rates in tension.................................. 327 E. Cadoni Design for Deconstruction .................................................................................................. 339 O. Hechler, O. P. Larsen & S. NielsenDemolition and recycling of demolition rubble after deconstruction ................................. 355 P. Kamrath & O. Hechler Part 2. Buildings Deconstruction of buildings: Masses and types .................................................................. 369 P. Kamrath Part 3. Bridges Demolition and deconstruction of bridges .......................................................................... 377 F. Portioli, M. DAniello, E. Cadoni, R.P. Borg & O. Hechler Conclusions......................................................................................................................... 390 F. Portioli & O. Hechler List of all contributions from COST C25393 Author Index397 iv vForeword ThispublicationistheFinalOutcomeoftheCOSTActionC25SustainabilityofConstructions: Integrated Approach to Life-time Structural Engineering which includes research reports, datasheets and guidelines in two Volumes. The Action has actively involved more than one hundred researchers, engineers and architects from 28 countries during four years since December 2006. TheActionhassummarizeditsresearchfieldbythewordingSustainabilityofConstructionsthat referstocreativecombinationofmethodsofstructuralengineeringwiththoseofsustainable construction.TheActionMembershavecollaborativelyexaminedtheories,methodsandtoolsfor cross-boundary and trans-disciplinary knowledge production, management and communication. These includeapproachestoassessenvironmental,socialandeconomicimpactsofconstructionactivities; methods to analyse and verify eco-efficiency of materials, components, buildings and infrastructures; andmethods of structuraldesign that incorporate holistic understanding of safety, eco-efficiency and sustainability. TheActionhasbeenascientific,socialandculturaljourneythathasgivenopportunitiesnotonlyto co-operateoverthebordersbutalsotolearnaboutvarioustraditions.TenManagementCommittee meetingshavebeenorganisedinninecountries,fromNorwaytoTurkeyandfromPortugalto Romania.Mostofthemhavebeenlargegatheringswithinvitedexpertsandmergedworkinggroup and team meetings. The events with true milestone results are the launching meeting in Eindhoven in December 2006, the First Workshop in Lisbon in 2007, the Mid-term Conference in Dresden in 2008, the Open Seminar in Naples in 2009 and the Second Workshop in Timisoara in 2009. The Action has alsoorganisedTrainingSchoolsinEindhovenin2008,inThessalonikiin2009andinVallettain 2010. A Student Competition was arranged in 2008-2009. TheActionhassuccessfullycontributedtothescientificadvancementofthemethodsoflife-time structuralengineeringandtotheimplementationofsustainableconstructionapproaches.Theideas and knowledge have matured throughout the Action, from the initial brainstorming of the proposal in 2005 until the production of this final outcome and organisation of the Final Conference in Innsbruck inFebruary2011.TheachievementsaremainlypublishedinfourProceedingsandTrainingSchool Books in relation to the Action events. Especially the following achievements are worth highlighting: - Sustainability assessment guidelines for both bridges and buildings - A methodology to assess sustainability of bridges - Integrated methodology for lifetime engineering including risk analysis and maintenance scenarios - Special Issue of Sustainability Journal with the Action Chair as a Guest Editor and several papers from the Members. TheActionwasorganisedinthreeWorkingGroupsinaccordancewiththemainresearchareas identified as necessary for the objectives: WG1 CriteriaforSustainableConstructions(topicslikeglobalmethodologies,assessment methods, global models and databases); WG2 Eco-efficiency (topics of eco-efficient use of resources in construction materials, products and processes); WG3 Life-timestructuralengineering(topicslike designfordurability,life-cycleperformance, including maintenance and deconstruction). The Action activities were taken care in several work packages in order to ascertain a good coverage of the most important topics as follows: WP1 State-of-the-art on LCA and LCCA methodologies as applied in participating countries WP2 Collection of information on databases of LCI and LCC for construction materials, construction products and processes and assessment of existing data and criteria WP3 Life-cycle performance: deterministic and stochastic simulation models WP4 Implementation of global methodologies for Sustainable Design and Building - case-study WP5 Identification and evaluation of existing and new functional materials, construction products and processes to comply with decrease of material use, decrease of waste, decrease of emissions and energy saving goals viWP6 Improvement of environmental performance of constructions (civil engineering structures, building structures and building envelopes). Improvement of the comfort in buildings (thermal, acoustic, lighting and quality of air), energy performance and the integration of innovative systems in buildings (mechanical, electrical and automation) WP7 Analysis of functional materials and applications and new technologies - case-study WP8 Life-cycle performance: verification methods for durability of constructions (degradation models and service design life) WP9 Monitoring of life-cycle performance (life-cycle safety, functionality, quality, demolition and deconstruction) WP10 Sustainable construction assessment and classification systems The Action piloted a project-like organisation of a scientific network by application of cross-cutting case-studieswhichaimedtosupportacoordinatedoutcome.Thecase-studieswerecontinuously reassessedandtheircomplexitywasincreasingfromessentiallybarestructuressuchasabridgeto completed buildings including structural parts and non-structural parts to allow a clear identification of allrelevantaspects.Bytheaidofcase-studies,thecurrenttechnologiesandmethodologieswere compared and initiatives for further developments were established. TheActionandtheWorkingGroupChairswanttowarmlythankallthecolleagueswhohave contributedtotheaccomplishmentsby writing,reading,debating and,nottheleast,byputtingupso much energy, mind and heart in hosting and organizing the C25 events. SpecialthankstoLuisSilvaformanagingtheActionwebsitewherealltheActionpublicationsare available and will be kept in coming years. A special gratitude is also addressed to Dr. Thierry Goger and Ms. Carmencita Malimban from COST OfficeandESF(EuropeanScienceFoundation)fortheirsupportandassistanceinadministrative matters. TheActionChairshopethattheachievementsoftheActionwillinspireEuropeanresearchersand practitioners to new efforts and innovations in the field of sustainable construction. December, 2010. The Chairs and the Working Group Chairs of COST Action C25, Action ChairAction Vice-chair Lus Bragana University of Minho Portugal Heli Koukkari VTT Technical Research Centre of Finland Finland WG1 ChairWG1 Vice-chair Rijk Blok University of Technology Eindhoven The Netherlands Helena Gervsio University of Coimbra Portugal WG2 ChairWG2 Vice-chair Milan Veljkovic Lule University of Technology Sweden Zbigniew Plewako Rzeszw University of Technology Poland WG3 ChairWG3 Vice-chair Raffaele Landolfo University of Naples Federico II Italy Viorel Ungureanu Politehnica University of Timisoara Romania COST Action C25 Webpage: www.cmm.pt/costc25/ viiPreface This book reports the outcomes of four years of activity carried out within the framework of the COSTActionC25bythemembersoftheWorkingGroup3(WG3):Life-timestructural engineering(designfordurability,life-cycleperformance,includingmaintenanceand deconstruction). The main objective of WG3 was the collection of methods and practices in the field of Life-timestructuralengineering,whichdealswiththeanalysisofstructuresaccordingtothebasic assumption that the performance is a time-dependant variable. Thedesignofstructuresbasedontheevaluationoflifecycleperformancerepresentsakey issue for the sustainable development of built-up areas, having the costs of construction sector a significant social and environmental impact on the world-wide economy. Theassessmentoflifecycleperformanceofstructuresinvolvesdifferentquestionsrelated bothtodesignandmanagementstages,including durabilityagainsttheeffectsof deterioration phenomena, monitoring and planning of maintenance operations. Asaconsequence,thedevelopmentofanintegratedapproachtolife-timestructural engineering entails the definition of design procedures based on performance levels considering mechanical safety as well as durability and obsolescence of structures.Inrecentyears,alargeamountofresearchwascarriedoutonthesetopics,butthe developmentofaholisticdesignapproach,basedontime-dependantperformanceand maintenance interventions, is still missing and far from application in current practice. For this reason,thecollectionofdesignmethodsandpracticespresentedinthisbookaimsat representingastepforwardinthedevelopmentofuser-orienteddesignapproachesandinthe dissemination of life-cycle analysis of structures. To cover the different aspects of lifetime structural engineering, WG3 has been organized in two Working Packages, namely WP8 and WP9. The main aim of WP8 was to analyze the different procedures for durability design based on semi-probabilistic and probabilistic approaches. In particular, its main tasks were the survey of thestate-of-the-artinverificationmethodsfordurabilityofconstructionsandtheanalysisof degradation models for prediction of service life of structures. According to performance based durability design, the safety of constructions is satisfied if the failure event corresponding to the maximum allowable value of degradation occurs after the design service life has expired, with a propersafetymargin.Inparticular,thecalculationoflife-timeofconstructionsrequiresthe identificationoftheenvironmentalloadsaffectingdurabilitythatwilllikelyactonthe structures. On the basis of the identification of environmental loads, the degradation factors and mechanismshavetobeevaluated.Oncedeteriorationmechanismsthatcouldactonthe structuresduringthelifecyclehavebeenidentified,correspondingdamagemodelshavetobe selected which provide degradation as a function of time on the basis of statistical formulations. Theprobabilisticcharacterizationofthedamagedevelopmentovertimeisnecessarysinceit influences the value of the safety factor to be considered in design phase. The definition of the maximumallowabledegradationvaluesfordifferentlimitstatesrepresentsthefinalstepof developmentofadurabilitydesignprocedure.Bothserviceabilityandultimatelimitstates relatedtobasicrequirementssuchasfunctionalityinuseandstructuralsafetyneedtobe defined.Withrespecttomechanicallimitstates,morespecificperformancelevelscanbe classified for durability design. In particular, serviceability limit states can be related to changes infunctionalityoraestheticsonthebasisofmaintainability,economyandenvironmental impacts.Finally,ultimatelimitstateshavetobedefinedforthedamageofstructuralmaterial which compromises the mechanical safety of constructions. WP9 was concerned with the monitoring of life-cycle performance and with maintenance and deconstructionofstructures,whichareotherimportantissuesfortheassessmentofresidual servicelifeandforthedefinitionofend-of-lifescenarios.Itsmaintaskswere:theanalysisof maintenance,repairandrehabilitationtechniquesandplanning;thesurveyandcondition assessment of structures in terms of safety and functionality; the classification and planning of demolition and deconstruction systems. Different techniques and numerical methods for damage monitoring of materials and evaluation of actual structural safety are available in the literature. The aim of monitoring analysis is to define the actions to avoid or defer obsolescence that is the viiiinabilitytosatisfythechangingoffunctionalandstructuralrequirements.Generallythese actions,suchasmaintenance,demolitionandrepairing,havethepurposeofminimisingthe impactsofobsolescencebeforecosts become substantial. In particular, the final objective is to reduce demolishing of facilities that have not reached their mechanical or durability limit states, andthuspromotethesustainabledevelopment.Inaperformancebasedapproach,thefactors and causes of obsolescence should be defined and identified, both in quantitative and qualitative terms,selectingdifferentlimitstatesforfunctionalityofdifferentstructuraltypologies.The different methods for decision making such as life cycle costing method and risk analysis have to be used to select the proper intervention strategy. Inthiscontext,thepresentbookcollectsaselectionofthecontributionspresentedbyWG3 membersandinvitedexpertsatthemeetingsandworkshops.Thecontributionshavebeen revisedonthebasisoftheotherpaperspresentedduringtheActionanddealingwithsimilar topics. The list of these reference papers is reported at the end of the book. ThebookisorganizedinfiveChapters,eachcorrespondingtooneoftheexpectedWG3 outputasitisreportedintheMemorandumofUnderstanding.Inparticular,thefirsttwo Chapters,correspondingtotheexpectedDeliverable8.1and8.2,representtheoutcomeofthe WP8,andtheremainingthreeChapters,matchingtheDeliverablesD9.1,D9.2andD9.3, summarise the activities carried out within WP9. The Deliverables were edited and revised with the contribution of E. Vesikari and O. Hechler, coordinators of WP8 and WP9, to whom I would like to express my gratitude for the excellent work they have done in the management of the research activity during these years.Finally,IwouldliketothankV.Ungureanu,vice-chairofWG3,andallthepeoplethat contributed actively to the achievement of the results presented in this book. Raffaele Landolfo Chairman of WG3 Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Introduction 1R. Landolfo University of Naples Federico II, Italy E. Vesikari Technical Research Centre of FinlandIntroduction This chapter reports the state-of-the-art on service life design methods and degradation models of building materials. It consists of 4 papers, first presented by the authors at seminars arranged byCOSTAction25SustainabilityofConstructions.IntegratedApproachtoLifetime StructuralEngineeringandlateronreviewedandexpanded.Thechapterisorganizedintwo parts. Part1isfocusedonCurrentMethodologiesforServiceLifePredictionandconsistsoftwo state-of-the-artpapers.Inthefirstpapertheprinciplesandmethodsbasedonexperimental investigation, experience from earlier constructions, theoretical calculations or a combination of these all, as referred to by Eurocode 0, are presented. In the second paper an introduction into theconceptsrelatedtoservicelifeplanningbasedoninternationalstandardization,isgiven. Especially the methods used in ISO 15686 series are addressed. Part 2 is devoted to Service Life Design Methods of Constructions and is also made up of two state-of-the-artpapers.Thefirstpaperdealswithservicelifedesignmethodsespeciallywith reference to bridges and civil engineering structures. The basic durability requirements reported inthestructuralcodes,aswellasfullprobabilisticproceduresthattakeintoaccountallbasic variableswiththeirstatisticalpropertiesareconsidered.Inparticular,thelimitingvalues method,thefactormethod(ISOSeries15686),thesemi-probabilisticdimensioningmethods suchasthelifetimesafetyfactormethodandthealternativeapproachesproposedintheFib Model Code on Service Life Design are introduced. The chapter is closed with a paper dealing with some aspects of risk based assessment of service life of building structures. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 2E. Vesikari Technical Research Centre of Finland R. Landolfo University of Naples Federico II, Naples, Italy Service Life Prediction Methods Referred to by Eurocode 0 1INTRODUCTION Eurocode 0, Basis of structural design, emphasizes the importance of durability and service life in todays structural design. Eurocode 0 pushes the development towards service life design by stating that the design working life should be specified and by imposing indicative service life requirementsfordifferentcategoriesofstructures.Forthedesignmethodsofservicelife Eurocode0doesnottakeapositionbutitstatesthatthedegreeofanydegradationmaybe estimatedonthebasisofcalculations,experimentalinvestigation,experiencefromearlier constructions, or a combination of these considerations. Whatever service life design methods are used in practical design degradation and service life data is needed. In this paper the methods of provision for reliable service life data are discussed. They may be categorized as follows: 1.Long-term aging tests; 2.Tests performed in laboratories as combined with long-term aging tests and; 3.Theoretical/analytic methods as combined with long-term aging tests and laboratory tests. 2LONG-TERM AGING TESTS Long-termagingtestsgeneratedataonlong-termageingunderin-useconditions.Long-term aging tests are necessary for obtaining reliable degradation and service life data in an objective way. Long-term aging tests can be divided into the following categories (ISO 15686-1): 1.Fieldexposuretests.Thisisthetraditionalwayofexposingmaterialspecimensor componentsonracksandstandsinspecificenvironmentalconditions.Largeseriesof materials can be tested in an exposure field. 2.Inspectionofoldbuildings.Theideaistousetheexistingbuildingstocktoacquire reliableservicelifedata.However,theusefulnessoftheinspectiondatafromold buildings may be limited by the difficulty of obtaining data on the history of the building andonthedifficultyindescribingtheinitialmaterialcharacteristicsandthe environmental conditions of the items observed. 3.Experimentalbuildings.Speciallyfortestingdesignedbuildingswhichcanexpose specific materials and components. 4.In-useexposuretests.Specimenscanbeexposedinbuildingsthathavenotbeen specificallydesignedforexperiment.Thisapproachisnecessarywheredegradationis directly related to user actions and behavior. To get more reliable data on degradation and service life of concrete structures new field tests havebeenstartedinNordicCountriesinrecentyears.Thepurposeofthesetestshasbeen provision for real data on the chloride penetration rate into concrete, corrosion on reinforcement Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Current Methodologies for Service Life Prediction 3andonfrostscaling both in marine and motorway conditions. Other field tests in non-chloride environmentshavebeenconductedforprovidingdataoncarbonation,corrosionof reinforcement and frost attack without salt. Figure 1. Field tests for concrete specimens beside a motorway near Kotka, Finland. Oneofthemostextensivelong-termexposureprogrammehasbeencarriedoutwithinthe International Cooperative Programme on Effects on Materials, including Historic and Cultural MonumentsintheframeworkoftheUNECEconventiononlongrangetransboundaryair pollution(Kucera2004).Theprogrammeaimedatevaluatingquantitativelytheeffectofair born acidifying pollutants on the atmospheric corrosion of several materials (metallic and stone materials,paintcoatings,electriccontactmaterials,glassmaterialsandpolymericmaterials.) The programme involves field exposure at 39 sites in 12 European countries and in the United States and in Canada (ISO/FDIS 15686-2, Annex A). Long-termagingtestsareaprerequisiteforreliabledegradationandservicelifedataas eventually only the data obtained by field tests can be considered as firm ground. As such the importanceofthefieldtestsisself-evident.However,theremaybeseriousproblemsin extendingtheapplicationofthesedatatoothermaterialsandtodifferentenvironmental conditions than those applied in the actual tests. To extend the application of the field test data beyondthelimitsoftheactualtestmayrequiresupplementarytestseitherinlaboratoryorin field. Therearealsoseriousproblemswithfieldtestsresultingfromthefactthatfieldtestsare long-lastingandexpensive.Thatiswhytheneedofservicelifedatacanneverbesatisfiedby fieldtestsonly.Othermethodsshouldbedevelopedtogeneratereliableservicelifedatamore quickly.Suchmorerapidmethodsmaybeacceleratedtestsperformedinlaboratoriesand applicationoftheoretical/analyticmodelingmethods.However,bothofthemmustbeverified and calibrated with observations from the field. 3TESTS PERFORMED IN LABORATORIES AS COMBINED WITH LONG-TERM AGING TESTS Togetdataonthedurabilityofmaterialsandstructuresacceleratedtestmethodshavebeen widely applied in different countries. Accordingly, many kinds of accelerated tests for material Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 4testingareavailable.However,onlyafewofthemareapplicabletoservicelifeevaluationas therelationshipbetweenthedegradationrateinthetestandtherealin-usedegradationrateis notknown.Thatiswhywhendevelopinglaboratorytestsforservicelifeevaluationthe degradationrateofthematerialorcomponentinin-useexposureconditionsshouldbestudied and related to the degradation rate in the test. The procedure which originally was proposed by RILEMTC71-PSLispresentedinISO15686-1andISO15686-2asslightlymodified.The methodology presented in ISO 15686-2 is shown in Figure 2. The ISO 15686 methodology contains the following phases: 1.Definition 2.Preparation 3.Pretesting 4.Aging exposure testing 5.Analysis and interpretation. Definition of the problem includes definition of the materials or structures to be studied, the rangeofthestudytobeestablished,andidentificationorspecificationofessentialdata. Essential data may be user needs, performance requirements, building context, exposure classes and materials characterization. Preparationphaseincludesidentificationofpossibledegradationmechanismsandagents (exposureconditions).Alsopostulationsaremadeonwhichperformancetestsareusedin observingthedegradationandonhowthedegradationcanbeacceleratedwithoutcausing undesired effects. When using accelerated short-term exposure care shall be taken to ensure that extreme acceleration does not result in degradation mechanisms that would not be experienced in service. Pretestingisbaseduponthepostulatesmadeabove.Bypretestingaconfirmationtothe degradationmechanisms,agents,selectedtestproceduresandperformancecharacteristicstests is provided. Testing of degradation is performed by observing and measuring 1.long-term degradation under in-use conditions (long-term-exposure) and 2.degradationunderexposureconditionsdesignedtoacceleratethedegradation(short-term-exposure). The ways to generate data on long-term aging under in-use conditions are the methods based onexperience(ref.chapter2).Short-termexposureiscarriedoutoveratimeperiod considerably shorter than the anticipated service-life. A short-term test may include operating at higher intensities of accelerating exposure or only be based on a shortened exposure period but withcontrolledexposureconditions.Acceleratedexposureisusuallydesignedfromthe information obtained from the pretests. IntheAnalysisandInterpretationphasestheresultsofthelong-term-exposuretestsare compared and related with the results of short-term tests. Consistencies and inconsistencies are usedtoincreaseunderstanding.Ifthereislittleornocorrelationbetweentheresults,thedata shouldbere-examinedandtheshort-termtestsmayneedtobealtered.Incaseofhigh consistencybetweentheresultsperformance-over-timefunctionsordose-responsefunctions can be formulated by experts so that the results of the short-term tests can be used in prediction of service life. Unfortunately very seldom accelerated tests are qualified for prediction of service life. There are many reasons for this (Pommersheim & Clifton 1985): 1.In a laboratory the specimens are usually tested in conditions substantially different from realoutdoorconditions.Asaresulttherelationshipbetweenthetimeinthetestand service life is obscured. 2.Factors affecting service life may be complex and not well known leading to biased test configurations in laboratory. 3.Accelerationofthetestmaychangethedegradationmechanismofamaterialfromthat found in in-service conditions. Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Current Methodologies for Service Life Prediction 5 Figure 2. Systematic methodology for service life prediction of building components (ISO 15686-2). Innormaloutdoorweatherthetemperaturesandmoistureconditionsvarywithinadayand withinayearverymuch.Thatiswhyitisalmostimpossibletoarrangeconditionsina laboratory that would completely correspond to the outdoor weather conditions. This is especially true for degradation types that are based on phase transitions such as frost damage.Innaturalweatherconditionsthepossiblefrostdamageofastructureishighly dependentonoccasionalperiodsofrainandsunshineleadingtoacomplexvariationof temperatureandmoisturecontentinsidethestructure.Thesamealsoappliesverymuchto chemical reactions such as carbonation and to electrochemical reactions such as corrosion. Thatiswhymostlaboratorytestscannotgiveafirmbasisforservicelifeprediction. However,laboratorytestsmaybeusefulastheygeneratedataonrelativedegradationrates between materials. Laboratory tests can also be used in studying the relative degradation rates at differenttemperaturesandmoisturecontents.Inairconditionedtestroomsofalaboratorywe canstudye.g.corrosionrateinconstanttemperatureandmoistureconditions,orcarbonation rateonconcreteinconstanttemperature,moisturecontentandCO2contentetc.Thisdatacan be utilised in the degradation modelling based on a theoretical/analytic framework. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 64THEORETICAL/ANALYTIC METHODS AS COMBINED WITH LABORATORY TESTS AND LONG-TERM AGING TESTS The theoretical/analytic methods of durability and service life can be grouped as follows: 1.Theoretical/analytic modeling 2.Computer simulation. Theoretical/analytic degradation modeling is based on general theories of diffusion, dissolution, convection, erosion, abrasion, chemical reaction, electrochemical corrosion, phase transition etc. Parameterstothetheoreticalframeworkareusuallysearchedbyexperimentalresearchbothin laboratory and in field. For a practically satisfactory model the parameters in the model should be those commonly used in the design of structures. Computer simulation refers to: 1.theoretical emulation of in-use climatic conditions, 2.determination of the temperature and moisture variations in a cross-section of a concrete structure, and 3.applicationoftemperatureandmoisturesensitivedegradationmodelssothatthe degradation over time and the service life can be predicted. 4.1Theoretical/analytic modeling Asanexampleoftheoretical/analyticmodelingofasimpledegradationmodelcarbonationof concreteischosen.Thefactthatthedepthofcarbonationisapproximatelyproportionaltothe square root of time can be theoretically derived by applying the diffusion theory with a moving boundary. In this theory the carbon dioxide is diffused through the already carbonized layer of concrete and reacts with the non-carbonated calcium minerals at the moving boundary, that is at the distance of Xc (depth of carbonation) from the surface of a structure. The carbon dioxide content between the surface and the moving boundary is assumed to be linear. Then the flux of carbon dioxide towards the moving boundary can be evaluated as: carbSXCD JA= (1)whereJisflux of carbon dioxide, kg/m2s, Dthe diffusion coefficient of concrete with respect to CO2, m2/s, Xcarbdistance of the moving carbonation boundary from the surface of the structure, m, ACS= Cs - Cxc, kgCO2/m3, CSCO2 content of air at the surface of concrete, kgCO2/m3, and CxcCO2 content of air at the moving boundary, kgCO2/m3. Thecarbondioxidefluxintoconcretemustbeinbalancewiththechemicalreactionratein concreteattheboundary.TheconsumptionrateofCO2dependsontheamountofnon-carbonated calcium in concrete: dtdXadtdQcarb carb=(2)whereQcarbisthe amount of bound carbon dioxide in concrete, kgCO2/m2 aCO2-binding capacity of concrete, kgCO2/m3 and ttime, s. BycombiningEquations1and2andintegratingovertime(XCarb=0whent=0),the following solution is obtained: at C DXSCarb A =2 (3) Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Current Methodologies for Service Life Prediction 7From Equation 3 it is seen that at constant conditions the depth of carbonation is proportional tothesquarerootoftime.IfACsisconstantinEquation3,theequationcan be presented in a more simplified form as: t k XCarb Carb = (4)where kCarb is coefficient of carbonation, mm/a0.5. Thecoefficientofcarbonationcanbestudiedexperimentallyinthelaboratoryatconstant moisture conditions and at constant temperature and for different mix designs of concrete. The real carbonation in in-use conditions can be studied by field tests. Another example of a typical degradation model is that developed for atmospheric corrosion in metal structures (Klinesmith, McCuen R. & Albrecht 2007). ) ( 201 1 ) (T T JH F DBeGClESOCTOWt A t y+|.|

\| + |.|

\| + |.|

\| = (5)whereyiscorrosion loss, m texposure time, years TOWtime of wetness, h/year SO2sulfur dioxide concentration, g/m3 Clchloride deposition rate, mg/m2/day Tair temperature, C. A, B, C, D, E, F, G, H, J and T0 are empirical parameters calibrated on the basis of the result of long term exposure tests. The power function of t can be theoretically developed. The rate of corrosionisdeclining(B (1) In the current practice, structural design is performed according to semi probabilistic and full probabilisticapproachconsideringdifferentlimitstateswhilefordurabilityonlygeneral principlesandapplicationrulesareprovided.Indeed,theISOStandard13823providesa generalframeworktoperformtheverificationofdurabilityatdifferentdesignlevels.Inthis section the approach developed in the ISO Standard 13823:2008 for the design of structures for durability is presented. The Standard has been published in May 2008 resulting in a very useful andcomprehensiveguideforthedesignofdurability.Structuressubjectedtoforeseeable environmental actions leading to a decrease of performances over time are covered. Mechanical actions producing deterioration are included, fatigue is neglected. Five informative annexes are included to provide illustrative examples on the procedures defined in the Standard (limit state method, the environmental actions causing degradation, the transfer mechanism from the macro-environmenttothestructurelevel,theenvironmentalactionsforstructuralmaterialsandtheir control and procedures for ensuring durability). For the prediction of service life, three different options are available. In particular, durability assessment shall be carried out according to past experience, modeling and testing. As for modeling, conceptual and mathematical modeling are defined. Durabilityisdefinedasthecapabilityofastructureoranycomponenttosatisfy,with plannedmaintenance,thedesignperformancerequirementsoveraspecifiedperiodunderthe influenceoftheenvironmentalactions,orasaresultofaself-agingprocessandforthe verification of durability the use of limit state method is recommended. The limit state method consistsmainlyintwosteps:thedefinitionandcharacterizationofrelevantdegradation mechanism that will likely act onto the structure and the verification of durability according to formula 1. The first step involves the characterization of the structure environment, the identification of thetransfermechanismandthedefinitionofenvironmentalactions.Inparticular,the environmentalactionthatcouldleadtoalossofperformance(damage,lossofresistance, displacements,aestheticlosses,etc.)istheresultoftheglobalenvironmentalconditionthat, throughatransfermechanism,transformstheatmosphericinfluencesinalocalattack.The structure environment is defined as the macro environment which the structure is located in and it comprises outdoor or indoor atmosphere, the ground or the water. It could be characterized by Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 20different parameters such as the humidity level, rainfall intensity, pollutants and so on. Gravity, capillarity,condensation,ventilationaresomeofthetransfermechanismscausing environmental actions. In Table 2 an example of the characterization of relevant deterioration mechanisms according to ISO 13823 is reported. Table 2. Example of characterization of relevant deterioration mechanisms according to ISO 13823. MATERIAL STRUCTURE ENVIRONMENT TRANSFER MECHANISM ENVIRONMENTAL ACTION ACTION EFFECT SteelOutdoor atmosphere Condensation; no drainage Atmospheric corrosion Thickness loss, aesthetic loss, rust expansion Reinforced concrete Outdoor/indoor atmosphere DiffusionChloride attack Loss of bond, failure of reinforcement WoodGroundDirect exposure Subterranean termites Loss of material, strength Regardless of the adopted level of the analysis, the verification of durability requirements can be performed either by service-life format or limit state format. Service life format consists in defining tD and calculate the predicted service life. Inafullprobabilisticapproachtheverificationofdurabilityconsistsinevaluatethe probability of failure and verify that it falls short of a target probability Ptarget (formula 2). The semi-probabilistic approach involve the definition of a characteristic value for the service life tsk andapartialsafetyfactorSaccordingtoformula3,whileadeterministicverificationof durability consists in evaluate the numerical value of the predicted service life tSP and check the basic requirements through formula 4. arg( )S D t etP t t P s s (2) skDStt> (3) S SP Dt t t = > (4) As for limit state format, three limit states can be defined, namely ultimate limit state (ULS) iftheresistanceRisequalorfallsshortoftheinternalmechanicalforce,serviceabilitylimit state(SLS)whenlocaldamageand/orrelativedisplacementaffectthefunctionalityand/orthe appearance of the construction and initiation limit state (ILS) corresponding to the initiation of significant deterioration. Alsointhiscase,theverificationshallbeperformedatdifferentlevelsbymeansof probabilityoffailureand/orpartialsafetyfactors.TheactioneffectSandtheresistanceRare regardedastimedependantvariablesand,inafullprobabilisticapproach,theyareboth modeled as random variables. The basic requirement for ULS is defined as: ( ) ( ) R t S t > (5) and the verification of durability can be formulated a s followings: | |arg ,( ) ( ) ( ) 0f t et ULSP t P R t S t P = s (6) Serviceability limit states refers to the functionality in use, an upper bound Slim for the action effect can be defined. The basic requirement and the verification according to a full probabilistic approach are expressed by Formula 7, 8: lim( ) S S t > (7) Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Service Life Design Methods of Constructions 21| |lim arg ,( ) ( ) 0f t et SLSP t P S S t P = s (8) In the current practice structural design is performed according to Level I methods (limit state analysis)whilefordurabilityonlygeneralprinciplesandapplicationrulesareprovided. Neverthelessadvancedmethodshavebeendeveloped.Inthefollowingsections,afterashort reviewoftheprinciplesandapplicationrulesgiveninthestructuralcodes,someexamplesof advanced procedures are presented. 3LITERATURE REVIEW OF SERVICE LIFE DESIGN METHODS 3.1Durability in structural design codes In this section the approach of the structural codes used in practice is reported. In general terms itispossibletosaythatdurabilityisdefinedasoneofthebasicrequirementtobeachieved during working life but usually only general provisions are recommended such as (JCCS 2001): choiceofsuitablematerialsthatwillnotdeteriorateduringthedesignworkinglife(ifthe maintenance is performed regularly). accounting for deterioration allowance in the dimensioning of structural elements. selecting a shorter working life for the elements that cannot be inspected and/or replaced. performing regular inspection and maintenance. The European Standard EN 1990:2002 Basis of structural design (EC0) addresses the general principlesandrequirementsforthedesignandverificationofstructuresaccordingtoalimit-state analysis. As for durability, Section 2 states that the structure shall be designed such that deteriorationoveritsdesignworkinglifedoesnotimpairtheperformanceofthestructure belowthatintended,havingdueregardtoitsenvironmentandtheanticipatedlevelof maintenance.Althougheachstructureshallbedesignedtohaveadequatedurability,only general provisions are recommended to meet these requirements (i.e. the use of preventative and protectivemeasures,par.2.2.5b,theproperchoiceofthedesignworkinglife,par.2.2.5e). Calculations,experimentalinvestigation,experiencefromearlierconstructions,ora combinationoftheseconsiderationsarerecommendedtoestimatethedegreeofany degradation. In case of steel structures, in EN 1993-1-1 (2004) only few common principles are stated for durabilityofmetalstructuresandinparticularforpreventingsteelbuildingsfrompossible causesofcorrosiondamage.ThecodereferstoEC0fordurabilityingeneralandgivessome recommendations such as the opportunity of providing corrosion protection measures by means of surface protection systems, improving the use of weathering or stainless steel and resorting to structuralredundancy.Noanalyticalmodelsand/orsafetyfactorstobeappliedinorderto achievetheserequirementsarespecified,thusthegeneralprovisiontobeappliedareleftto practitioner experience and expert judgment. Furthermore,itcanalsobenotedthataccordingtoEC0,themechanicaldesignand verification of structures shall be performed by limit state analysis using partial safety factors. A design is considered to be sufficient if the relevant limit states are not reached when the design values of the variables are introduced into the analysis. Thedesignvaluesarederivedfromrepresentativevalues(characteristicornominalvalues) throughtheuseofpartialsafetyfactors.AsfortheresistanceRthedesignvalueRdcanbe derived from characteristic value Rk through the partial safety factor M which take into account theuncertaintiesrelatedtomaterialproperties,geometricaldeviationfromnominalvaluesand modeluncertainties,theeffectofdeteriorationontheresistanceisneglected.Inaddition,itis worthtoemphasizethateithertheactionsonthestructureandtheresistanceitselfaretime-dependantvariablesbutthetimeisusuallyneglectedinordinarydesignthankstospecific assumptions.Bytheresistanceside,thebasicassumptionthatallowstoneglectthevariable timeisthatthemaintenanceactionwillensuretheadequateresistancelevelasatin-built condition time. Thus, if planned maintenance actions are inadequate or not performed at all, the reliability of the structure over time could fall short of the target reliability level defined at the design stage. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 22Onlyfatigueisregardedastime-dependantphenomenabutalsointhiscasethedegradation ofresistanceovertimeisneglected.Incaseofsteelstructuresitisassumedthatthefatigue strengthreferstostructuresoperatingundernormal atmospheric conditions and with sufficient corrosion protection and regular maintenance (Eurocode 1993-1-9:2005, par. 1.1.7). IntheEuropeanStandardEN206-1:2002requirementsforconcretetowithstandthe environmental actions are given either in terms of limiting values for concrete compositions and establishedpropertiesorrequirementsmaybederivedfromperformance-relateddesign methods.Therequirementsshalltakeintoaccounttheintendedworkinglifeoftheconcrete structure. The performance-related design methods according to the following: a)Therefinementofthelimitingvaluesmethodbasedonlong-termexperienceoflocal materials and practices, and on detailed knowledge of the local environment. b)Methodsbased on approved and proven tests that are representative of actual conditions and have approved performance criteria. c)Methodsbasedonanalyticalmodelsthathavebeencalibratedagainsttestdata representative of actual conditions in practice. 3.2The Limiting values method The limiting values method refers to limiting values for the composition of materials and proven properties of materials and structural measures in various classes of exposure. By applying the limitingvaluesastructuraldesignercanprovethestructuretofulfiltherequirementsofthe intended working life. An example of the limiting values method is given in EN 206-1:2002. A recommendation for the choice of limiting values for concrete composition and properties is given in Table 3 when using CEM I cement. The values in Table 3 are based on the assumption of an intended working life of the structure of 50 years. Table3.Exampleofalimitingvaluesmethod.Recommendedlimitingvaluesforcompositionand properties of concrete according to EN 206-1:2002. TheabbreviationsXC,XS,XD,XFandXArefertoexposureclasseswhicharedefinedin thesamestandard.ThelimitingvaluesofTable3areinformative,meaningthattheycanbe revised on national bases. The minimum values for concrete cover are given elsewhere. 3.3The Factor Method The factor method (ISO 15686-1 2000) belongs to level 0 design procedure, the assessment and evaluation of service life is deterministic and it is based on the identification of the main factors influencing service life. Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Service Life Design Methods of Constructions 23Itisclearlyspecifiedthatthemethoddoesnotprovideaguaranteeonservicelife,butit simply gives an estimation on the long lasting of components under certain condition. The expected numerical value of service life (ESL) is obtained by adjusting the value of the reference service life (RSL), through series of factors according to Formula 9: ESL RSL A B C D E F G = (9) where:A.Quality of components as supplied to the project; B.Design level of a component or assemblys installation; C.Work execution level or skill level of the installers; D.Indoor environment; E.Outdoor environment; F.In-use conditions; G.Maintenance level. For the evaluation of the ESL, the RSL as well as the factors A - G are required as input. The RSLshallbeadocumentedperiodoftimeinyearsbasedondataprovidedbytests,building codesandexpertjudgments.Asfortheadjustmentfactors,thestandarddoesnotprovide explicit values. The designer has to establish their numerical values (CIB 2004b). 3.4The Life time safety factor method ThelifetimesafetyfactormethodindurabilitydesignwasfirsttimepresentedintheRILEM Report on concrete structures (Sarja & Vesikari 1996) and then developed within the framework oftheEUProjectLIFECON(Sarja2004).Thismethodrepresentsafirstattempttoapply simplified probabilistic method to everyday engineering design procedures. Themethodisusedforthecalculationofdesignlifeofconstructionsanditisbasedon probabilistic degradation models, which consider the decrease of structural resistance caused by differentclassesofenvironmentalloads.Thedurabilitydesignbasedonlifetimesafetyfactor methodisanalogouswiththestaticlimitstatedesign.Inparticular,itisrelatedtocontrolthe failure probability by considering the effects on R of the environmental loads acting during the entire life time cycle, while static limit state design is devoted to control the structural reliability ofconstructionsunderexternalmechanicalloading.Indurabilitylimitstatedesignthe resistanceRisconsideredasatimedependantvariable,contrarytostaticlimitstatedesign, where the effects of time are usually neglected for R (Figure 1). In particular, a deterioration function D can be formulated on the basis of the time dependant resistance R according to formula 10: ( ) (0) ( ) D t R R t = (10) where D(t) is the deterioration at time t; R(0), R(t) are respectively the resistance at t = 0 and at the generic time t of the life cycle.

Figure 1. Representation of the life time safety factor design approach by R, S, t and D, t variables. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 24Becauseofthedifferentsourcesofuncertaintiesthatareinvolvedinthedefinitionofthe variation of capacity with time, the values in the previous equation are usually taken as mean or characteristicvalues.TheloadSisusuallyadoptedtobeconstantwithtime,anditsdesign value is also taken as a mean or characteristic value, multiplied by the relevant safety factors. The failure event corresponds to time tmax when the capacity R is equal to load S. The difference R(t)-S represents the reduction with time of the safety margin: max( ) R t S = (11) Onthebasisofpreviousconsiderations,adurabilitydesignprocedureorganizedinto different steps can be formulated, as specified in the following. First,thetargetservicelifetghastobedefinedfortheconsideredconstructions.Thetarget service life corresponds to the design life defined in section 2. The reference value of the target service life has to be selected according to mechanical design procedure. Once the reference period has been stated, it is possible to identify the environmental loads S thatwilllikelyactontostructure.Eachenvironmentalloadisanalysedandquantified,where relevant, in a statistic way. The analysis of the environmental condition has to be performed in ordertodefinetheprojectbackground.Onthebasisoftheidentificationofenvironmental loads, the degradation factors and mechanisms should be evaluated. Once deterioration mechanisms that could act onto structures during the life cycle have been identified, corresponding damage curves should be considered as a function of time in the form: ( )nmD t t o = (12) whereDm(t)isthe mean value of degradation isa constant coefficient ttime ndegradation mode coefficient. Substituting tmax in eq. (12), we have: max max( )mD t D = (13) where Dmax = R(0) - R(tmax) = R(0) - S represents the maximum allowable value of degradation (e.g. the maximum allowable mass loss and/or corrosion depth in the cross section of a beam). Durabilityrequirementsarefulfilledifthefailureevent(13)occursafterthedesignservice life had expired, with a proper safety margin. That could be expressed according to formula 14: max0d gttt t= > (14) wheretdisthe (design) service life tmaxisthe calculated mean value of the service life corresponding to D(tmax) = Dmax t0isthe central lifetime safety factor tgisthe target service life. AssumingthatdegradationisnormallydistributedandthestandarddeviationofDis proportionaltothemeandegradation,thecoefficientofvariationVDbeingconstant,itcanbe shown that the central lifetime safety factor of the design life depends only on safety reliability index, the coefficient of variation of D and the exponent n, according to the formula 15: ( )101nt DV | = + (15) On the basis of such assumptions, the safety reliability index depends on both the maximum allowablefailureprobabilityfortheselectedlimitstateandthedegradationfunction.In particular,itisafunctionofthemeanvaluesandstandarddeviationsofRandS,asfollows (formula 16): Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Service Life Design Methods of Constructions 252 2R SR S |o o=+(16) 3.5The fib model code for service life The fib was founded in 1998 from the unification of the Euro-International Concrete Committee (CEB,ComitEuro-InternationalduBton)andtheInternationalFederationforPrestressing (FIP, Fdration Internationale de la Prcontrainte). Among different tasks related to structural designofconcretestructures,durabilityconcernshavebeendeeplyinvestigatedsince1978 (CEBBulletin148Durabilityofconcretestructures1982,Bulletin182Durableconcrete structures 1989 and Bulletin 238 New approach to durability design 1997). ThelatestreportModelCodeforServicelifeDesign(MC-SLD,fib2006)wasprepared withinthefibTaskGroup5.6withtheaimtoprovideabasisandageneralframeworkfor standardization of durability based design approach. It is worth to point out that the code refers only to concrete structures, other construction materials are not considered thus far. TheaimofMC-SLDistoprovideadesignapproachfordurabilitysimilartomechanical design. To this end the environmental actions and the resistance of concrete against this action aremodeledquantitatively.Thedesignapproachcouldbesummarizedinfoursteps.Thefirst stepconsistsinthedefinitionofthedeteriorationmechanismaffectingthedurabilityofthe structureintermsofquantitativemodels.Thusthedurabilityperformanceshavetobedefined by means of limit states, the likelihood of attaining the defined limit states has to be evaluated and finally the classification of the type of limit state by means of ultimate and/or serviceability limitstatesshallbeprovided.AccordingtoMC-SLD,fullprobabilistic,partialfactor,deemed to satisfy design and avoidance of deterioration design criteria are established (Figure 2). Figure 2. MC-SLD, an extract of the flow chart for service life design. Full probabilistic approach is suggested for exceptional structures and in order to apply a full probabilisticprocedure,validatedprobabilisticmodelshouldbeavailableandthemodel parametersaswellastherelateduncertaintiesshallbequantifiable.Inthepartialfactor approachtheprobabilisticnatureofthevariablesistakenintoaccountbymeansofpartial factors.Deemedtosatisfymethodsconsistsingivingasetofrulesbasedonstatistical evaluationofexperimental data and field observation and/or on a long term experience for the dimensioning,theselectionofmaterialandproductandexecutionprocedures.Specific requirements to the workmanship, concrete composition, cover thickness etc. are recommended. Theavoidanceofdeteriorationmethodconsistsinadoptingsolutionandmaterialsothat deteriorationwillnottakeplace(i.e.the adoption of membranes to separate the structure from theenvironment,theuseofnonreactivematerialetc.).Furtherdetailsoftheprocedure developed in the MC-SLD can be found elsewhere (fib 2006). Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 264CONCLUDING REMARKS Theproblemofdurabilitycanbedescribedbytwobasicvariables,thecapacityRofthe structure to withstand the deterioration that will likely occur during service life and the loads S actingontothestructureduringservicelife.Itshallbeverifiedthatduringservicelifethe capacity meets or exceed the demand onto the structure. In general both the capacity R and the loadSaretimedependantvariables.Inthemechanicaldesignusuallythetimeisneglected thanks to specific assumptions, but in durability design the time is explicitly considered on the resistance side and the decrease of resistance over time is calculated. As for traditional structural design, different approaches can be used to evaluate the reliability against durability, namely: Prescriptive approach (instructions and design rules are provided) Deterministic approach (level 0) Semi-probabilistic approach (partial safety factors are defined, level I methods) Probabilistic approach (reliability structural analysis is performed, level III and II) Inthecommonpractice,structuraldesignisperformedaccordinglevelImethodswhilefor durabilityonlydesignrulesandgeneralprovisionsareestablished.Indeed,severaldesign approaches have been developed in the framework of international research. The publication of theISOStandard13823:2008representsafirststeptowardsthecodificationofdurability design. The Standard does not address a specific design procedure for durability as no specific model for the degradation of construction material neither numerical or reference values for the partial safety factors to be applied are provided but it represents a first and important attempt to createageneralframeworkandaunificationroleinthefieldofverificationofdurability. Currentlythereisagapbetweenresearchersandpractitionersinthefieldofdurabilityand additionaleffortshavetobeprovidedtointroduceadvancedservicelifedesignmethodsinto day to day engineering practice. REFERENCES Architectural Institute of Japan 1993. The English Edition of Principal Guide for Service Life Planning of Buildings, Architectural Institute of Japan, Japan. BritishStandardsInstitution1992.BS7543:1992GuidetoDurabilityofBuildingsandBuilding Elements, Products and Components. British Standards Institution, London, UK. CanadianStandardsAssociation1995.CSAS478-1995Guidelineondurabilityinbuildings.Ottawa, Canada. CIB-Council for Research and Innovation in Building and Construction 1999. Agenda21 on Sustainable Construction. CIB Report Publication 237. Rotterdam. CIB-CouncilforResearchandInnovationinBuildingandConstruction2004a.PerformanceBased Methods for Service Life Prediction State of the Art Reports. Part A, Part B. CIB Report Publication 294. CIB-Council for Research and Innovation in Building and Construction 2004b. Guide and Bibliography toServiceLifeandDurabilityResearchforBuildingMaterialsandComponents.CIBReport Publication 295. CEN 2001. EN 1990. Eurocode. Basis of structural design. CEN 2004. EN 1993-1-1. Eurocode3: Design of steel structures. Part 1-1. CEN 2002. EN 206-1. Concrete - Part 1: Specification, performance, production and conformity. EuropeanUnion1988:TheConstructionproductsdirective.CouncilDirective89/106/EECof21 December1988ontheapproximationoflaws,regulationsandadministrativeprovisionsofthe Member States relating to construction products, European Union, Brussels, Belgium, December. Fib-Fdration Internationale du Bton 2006. Model Code for service life design Bulletin 34. Stuttgard. ISO-InternationalOrganizationforStandardization2000.ISO15686-1:2000,Buildingandconstructed assets- Service life planning- Part 1: General principles. ISO-International Organization for Standardization 2008. ISO 13823:2008(E). General principles on the design of structures for durability. EuropeanCommission1999:DurabilityandtheConstructionProductsDirective,GuidancePaperF, European Commission, DG III, Brussels, Belgium. JCCS-JointCommitteeonStructuralSafety2001.ProbabilisticmodelcodePartI.availableon www.jcss.ethz.ch. Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Service Life Design Methods of Constructions 27Sarja,A.2004.LifecondeliverableD2.1.Reliabilitybasedmethodologyforlifetimemanagementof structures. Finland. Sarja,A.&Vesikari,E.1996.Durabilitydesignofconcretestructures.RILEMReportofTC130-CSL.RILEM Report Series 14. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 28Sz. Wolinski Rzeszow University of Technology, Rzeszow, Poland Risk Based Approach to Service Life Assessment of Building Structures 1INTRODUCTION Building structures should be designed in such a way that they will sustain all actions likely to occurduringexecutionandintendedservicelife,withappropriatedegreeofreliability,in economicalwayandtakingintoconsiderationenvironmentalaspects.Thereliabilityrequired forstructurescanbeachievedbydesigninaccordancewithcurrentstandards,appropriate execution and quality management measures (EN 1990, 2002, ISO 2394, 1998). The difference betweendesignofanewstructureandassessmentofanexistingstructureconsistsinthe availableorcollectableinformationaboutthestructurerelatedtomaterials,actionsandthe structuralperformancecharacteristics.Assessingofexistingstructurescanbeunderstoodasa process of model formulation, consequence evaluation and model updating by introducing new information (Faber at al., 2006, Wolinski, 2006). Anexistingstructureneedsanassessmentinsituationswhere:unexpectedorexcessive degradationhasbeenobserved,thestructurehasbeensubjectedtonon-foreseeneventsor increased loading, modification in use or increased service life is planned,the structure has not been inspected for a long period of time, etc. A structure can be measured, inspected, tested and proof-loaded.Allinformationrelevantforassessingtheperformanceofastructureshouldbe collected, but at a cost. Currentdesigncodesarebasedonthelimitstates concept and the reliability based methods are recommended to provide a minimum safety level of structures. The measure of reliability is identified with the probability of survival that depends on safety of people and balance between the costs of failure and safety measures. Calibration of reliability levels is generally based on the pastexperienceandtargetvaluesofthesurvivalprobabilitydependsonvariousdesign situations, failure modes, reference periods and so on, are generally notional numbers. Recently thereisgenerallyrecognitionthatthestructuraldesignandassessmentshouldinvolvethe considerations of reliability, safety, economic and environmental aspects and that these aspects shouldbeaddressedexplicitlyandfullyquantitatively.Acceptancecriteriainstructural engineeringshouldtakeintoconsiderationsalsodifferentconsequencesoffailure,especially: lossofhumanlifeorseriousinjuries,economicalandenvironmentallosses.Theycanbe derived from social indicators, which are statistics that express some aspects of quality of life in society. The reliability based criteria that take into consideration only the probability of failure and loss of functionality does not meet these requirements. Thepresentpaperdealswithrecentdevelopmentsofservicelifeassessmentofbuilding structuresbasedonthegeneralmethodsforassessingriskofsystemsconstructedbystandard components.Theapproachpresentedutilizesquantitativedefinitionsofexposure,robustness andvulnerabilityofbuildingstructuresandtakingintoconsiderationboth,directandindirect Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Service Life Design Methods of Constructions 29consequencesoftheirfailureandlossoffunctionality.Thepresentedapproachisillustrated through an example application. 2RISK INDICATORS 2.1Hazard and risk in structural engineering Hazard is defined as an attribute of activities which may cause harm to persons and assets or as asetofcircumstancewiththepotentialforcausingeventswithundesirableconsequences (Vrouwenvelderetal.,2001).However,intheEurocodehazardisdefinedasanunusualand severeevent.Thetwomajorcategoriesofhazardinvolvedinthebuildingprocesscanbe distinguishedonthebasisoftheirnature,namelynaturalhazardsandman-madehazards. Natural hazards resulting from variations of structural materials and products properties, actions appliedtoastructure,geometricaldata,whicharemodeledasstochasticvariableswith correspondingstatisticaldistributionsandparameters.Man-madehazardsincludeuncertainties duetounintentionalandintentionaldeparturefromtheacceptedpracticeandverified procedures(humanerrors),newmaterialsandtypesofstructures,innovationsindesignand construction, new technologies and methods of execution, new or modified models for structural analysis and dimensioning, fires, explosions and other severe events caused by man who are not involvedinthebuildingprocessbutalsothepressureonthedesignersduetotheshortageof time, money and the political climate. Negative consequences are defined as a possible outcome of desired or undesired events that maybeexpressedquantitativelyorqualitativelyintermsofpersonalinjury,death,monetary loss, environmental and social damage. Risk may be referred to as a measure of the danger or hazard that undesired events represents forpeople,economyandenvironment,andisdefinedasacombination(usuallyaproduct)of theprobabilityofoccurrenceandtheconsequenceofa specified hazardous or undesired event (Steward & Melchers, 1997, Vrouwenvelder et al., 2001). For a set of hazardous design situation iHthe total risk R can be calculated as follows (Faber et al., 2006): = = ==H D Sninjnkk j k i j iS C D S p H D p H p R1 1 1) ( ) ( ) ( ) ( (1) wherethestructureissubjectedto Hn differenthazardsthatmaydamagethestructureinDndifferent ways and the performance of the damage structure can be discretised into Snadverse statesSkwithcorrespondingconsequencesC(Sk),andp(Hi)istheprobabilityofoccurrenceof thei-thhazard iH ,) (i jH D p istheconditionalprobabilityofthej-thdamagestateofthe structuregiveninthei-thhazardand) (j kD S p istheconditionalprobabilityofthek-th adverse overall structural performance S given in the i-th damage state. 2.2Exposure and consequences Methodsforassessingriskofsystemscanbeefficientlyappliedforassessingservicelifeof buildingstructures.Generally,abuildingstructurecanbedefinedasaboundedgroupof interrelated,interdependentorinteractingelementsforminganentitythatachievesadefined objective in its environment through interaction of its parts. A building structure considered as the system involves structural system formed from a physical subsystem, environmental, social and human subsystems. Anexposureisconsideredasanyevent or set of events with the potential to cause damage, loss of functionality or failure to a system. A description of a series of events in time and space, andinter-relationshipamongtheevents,providedtheoccurrenceofahazardiscalleda scenario.Exposureeventscouldcomefromdifferenttypesofdirectandindirectactionsor environmental processes, errors and other disturbances. In the risk based approach conventional actionorloadcombinationmodelsestablishedinreliability-baseddesignmethodsshouldbe replacedwitheventscenarios,forexampleintheformofeventtree.Theeventtree Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 30representationisatoolforevaluatingeventscenariosthatcouldoccurtothestructure,andit also incorporates the associated probabilities of occurrence (Fig. 1). Twosignificanttypesofconsequences,bothimmediateandariseafteracertaintime, associated with the exposure to a structural system may be distinguished: -direct consequences induced by damage to the individual constituents of the system, -indirect consequences beyond the direct consequences, induced by changes of the system. C DIRECT RISK F EXk C D INDIRECT RISK FINDIRECT RISK SDi EXj DNO RISK Figure1.Representationoftheexposureinformofaneventtree.Clarification:SDi-theith versionof structuredesign,EXj-thejthscenarioofexposure,D-damage,D -nodamage,F-failure,F -no failure, EXk - the kth extraordinary event, C - collapse,C- no collapse. 2.3Direct and indirect risk The risk RD associated with all direct consequences due to exposure events may be assessed as follows (Faber et al., 2006): ==EX CDnknlk l D k l DEX p C c EX C p R1) ( ) ( ) ( (2) wherenEXisanumberofexposureevents,nCDisanumberofpossibledifferentstatesofall constituentsoftheelementCl,) (k lEX C p istheconditionalprobabilityofthel-thdamage stateoftheelementClontheexposureeventEXkwithprobabilisticcharacterization) (kEX pand) (l DC cis the direct consequence associated with the l-th of nCD possible state of damage of all constituents of the element Cl. The risk RID due to all indirect consequences of exposure events may be calculated using the formula: = = ==EX CD STnknlnmk k l k l m l D m ID IDEX p EX C p EX C S p C c S c R1 1 1) ( ) ( ) , ( )) ( , ( (3) where STn isanumberofpossibledifferentstructurestatesSmassociatedwithindirect consequences)) ( , (l D m IDC c S c and) , (k l mEX C S p istheconditionalprobabilityofindirect consequences on a given state of the constituents Cl and the exposure EXk. Following(Faberetal.,2006)thedirectriskRDisthemeasureofsystemvulnerabilityand the ratio of direct risk to the total risk (R = RD + RID) is the measure of system robustness. For a given set of the N different scenarios of actions, the risk associated with each scenario canbecomputedandthecorrespondingindexofrobustnesscanbecalculatedusingthe generalization of equation (Faber, 2007a, Faber et al., 2007b): Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Service Life Design Methods of Constructions 31 = ==+=Nii IDNii DNii DRR RRI1,1,1,(4) Therobustnessindextakesvalues1 0 s sRI dependingonthesourceofrisk.Ifallriskis duetodirectconsequences,thesystemiscompletelyrobustand1 =RI .Ifallriskisdueto indirect consequences, the structural system has no robustness and0 =RI . 3RISK ACCEPTANCE Acceptancecriteriainstructuralengineeringshouldtakeintoconsiderationsdifferent consequencesoffailure,especially:lossofhumanlifeorseriousinjuries,economical,social andenvironmentallosses.Theycanbederivedfromsocialindicators,whicharestatisticsthat expresssomeaspectsofqualityoflifeinsociety.Whendiscussingtheproblemofacceptable risklevelsitisnecessarytotakeintoaccountthefactthatindividualsmayhaveadifferent viewpointtowhatisacceptableriskascomparedtotheviewpointofthesocietyanddecision makers.Consideringactivitiesrelatedtocivilengineeringitisobviousthatthepreferencesof individualsmaybeincontradictionwiththepreferencesofsociety.Therefore,anormative approachtomodellingofpreferencesandcriteriaforriskacceptanceareessential.Commonly used formats for risk acceptance are the so-called FN-diagram or FN- matrixes. Figure 2. FN-diagram according to ISO 2394: 1998. InFigure2,theFN-diagramforacceptablerisktohumanlifeaccordingtotheformula: o < N A year f P ) / ( ,fromtheISO2394ispresented.Onthex-axistheconsequencesin terms of the number of fatalities are given, and on the y-axis the probability of occurrence of the collapseeventsaregiven.TheplotsinFigure2showdomainsofunacceptable,tolerableand negligible risk to human life characterized by lines corresponding to constants: A = 0.1,2 = oand A = 0.01,2 = o(Wolinski, 2008). Inmanycasesrelativecostsofsafetymeasuresandconsequencesofstructuralfailuresare definedbymeansoflinguisticvariablesinnaturallanguage.Forinstance,intheISO2394: 1998standardthetargetprobabilitiesoffailurearereferredtotherelativecostsofsafety measuresdefinedas:high,moderateandlow,andconsequencesofstructuralfailureas:great, moderate, some and small. InFigure3consequencesoffailureexpressedbymeansofthefuzzytargetprobabilitiesof failureandtherelativecostsoffailuredefinedintermsofthemembershipfunctions,are presentedtogetherwiththecorrespondingfrequency-consequencesdiagramcorrespondingto recommendations given in ISO 2394: 1998 and EN 1990: 2002. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 32 Figure 3. Fuzzy frequency-consequences FN-diagram. Complexsocialindicatorsarecommonlyusedforoptimizationofthetolerablerisk. Unfortunately,sufficientlyreliableeconomicanddemographicdataaswellasactualfailure ratesfordifferenttypesofbuildingandengineeringstructuresare insufficient, time-dependent and generally questionable. Moreover, life quality has more dimensions than consumption, long life and leisure time which are usually considered. Ethical, esthetic, cultural and religious values can not be measured by the standard life quality indicators, that usually incorporate questionable orunreliabledatarelatedtothequalityoflife:costofavertingtofatality,lifeexpectancy, personal well-being, time for leisure, healthy ecological environment, cultural heritage, etc. 4RISK BASED ASSESSMENT OF SERVICE LIFE Thedifferencebetweendesignofanewstructureandassessmentofanexistingstructure consistsintheavailableorcollectableinformationaboutthestructurerelatedtoexposure, vulnerability,robustnessandcostsofimprovementsofthestructuralperformance characteristics.Theassessmentofservicelifeofexistingstructuresmaybeunderstoodasa process of model formulation, consequence evaluation and model updating by introducing new information.Thefollowingmainstepsintherisk-basedframeworkoftheassessmentprocess may be distinguished: -Selectalistofdoubtsaboutastructure,anexposure,relevantlimit states (deterioration, ultimate and serviceability limit states), reference period, etc. -Confirmdoubtsbymeansofthepreliminaryinspection(studyofdocumentation,site visit, simple check, etc.). -Ifdoubtsareconfirmed,thenfurtherinspection,investigationsandanalysisshouldbe undertaken which include steps as follows: -formulation of prior uncertainty models referred to actions, deterioration mechanisms, behaviour of elements and the structure and models for the service life prediction, Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Service Life Design Methods of Constructions 33-formulationoftime-dependentperformanceandlimitstatefunctions,establishing posteriorprobabilisticmodels(usingdifferentframeworks:conventionalorBayesian or fuzzy-statistical), -calculating direct and indirect risk, an index of robustness and remaining service life. -Check compliance with codes and regulations (vulnerability), the level of robustness and remaining service life of the structure. -If the results of checking are positive, update the maintenance strategy. -If the results of checking are negative, choices of repair and strengthening or demolition of the structure is necessary, taking into account: -assumed period of the remaining service life, -refined limit states analysis, -direct and indirect risk after repairs and strengthening, -economical decision analysis. 5ILLUSTRATIVE EXAMPLE Anexistingopen-areasupportstructureforwarehousecraneconsistsofn=215=30 prefabricatedreinforcedconcretebeamssimplysupportedoncolumnsfixedinisolatedpad foundations.Thestructureisaged22years.Due to exposure conditions (i.e. large variation of moistureandtemperatureaswellasfreezeandthaw)deteriorationsuchasrebarcorrosion, crackingandspallingordelaminationofconcreteandtheprematurelossoffunctionalityor collapse of beams can occur. -Thepreliminaryinspectionhasshowntheappearanceofmoistureandruststains,few steep cracks and poor quality of concrete. -Considered beams have been designed using the limit states method and a partial factors format.ThedesignvaluesloadbearingcapacityforbendingisMd=224kNm. Correspondingvalueofthenotionalfailureprobabilityofonebeamisequal 610 5 =fp .Thestructuralsystemconsistsof30beamsandformstheseriessystem. For series system with perfectly correlated elements1 =ij , 6,10 5 = =f i fp pwhile forseriessystemwithuncorrelatedelements0 =ij ,. 10 5 . 1 ) 1 ( 14, = =ni f fp pWhen elements of the considered series system are equally correlated and the coefficient ofcorrelationis80 . 0 =ij ,theprobabilityoffailureequals. 10 1 . 25 ~fpUnfortunately, values of correlation coefficients between the resistance of different beams are extremely uncertain. -Fivestagesindevelopmentofreinforcementcorrosioninconcretearecommonly distinguished (Bertolini et al., 2004, fib, 2006): depassivation of concrete cover, initiation ofreinforcementcorrosion,crackingofconcrete,spallingofconcretecover,endof service life or collapse. -UsingsimplifiedformulasbasedontheFickslawofdiffusionandassumingthatthe increaseinthevolumeofthecrackisequaltothevolumeofthecorrosionproducts produceswhenthediameterofthereinforcementbarisreduced,thecorrosioninitiation timetcor,thecrackinginitiationtimetcrc,thespallinginitiationtimetsplandthetimeto collapse tcol were calculated (Thoft-Christensen, 2000, fib Bulletin, 2006, Liu & Weyers, 1998).TheMonteCarloMethodwasusedtoperformtheprobabilisticcalculations. Results of these calculations are presented in Figure 4. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 34 Figure4.Densityfunctionsfor:(a)corrosioninitiationtimetcor,(b)crackinginitiationtimetcrc,(c) spallinginitiationtimetspl,(d)timetocollapsetcol;wheretistimeinyears, (...)t ismeanvalue, 01 . 0 )~(~= s col colt t p t -Diagrams of theoretical and updated relationships between the age and the relative value of direct consequences are presented in Fig. 5. According to the results of inspection after 22yearsofservicethecostofnecessaryrepairsandprotectionofdamagedbeamswas estimated at 5% of all beams replacement, instead of calculated theoretically 2%. In order to correct the theoretical diagram its second part was shifted up in a parallel direction and the updated mean values of corrosion and cracking initiation time were calculated: cort= 14.4yearsand crct =23years.Thennexttwopartsofthetheoreticaldiagramwere shifted up in the same way and the updated mean value of spalling time splt= 38.8 years and time to collapse colt= 73.2 years were assessed. Figure 5. Diagrams of theoretical and updated relationships between the age and direct consequences. Chapter 1. State-of-the-Art Report on Service Life Prediction and Design Methodologies: Service Life Design Methods of Constructions 35-Service life of the structure may be defined in two ways: when the relative value of direct riskRDachievesthetargetvalue(economiccriterion)orwhentherelativevalueof indirect risk RID due to collapse caused by corrosion of tension reinforcement achieves the targetvalue(humansafetycriterion,forinstanceaccordingtoISO2394:1998). Replacementcostofall30beamsmakesup100%ofdirectconsequencesandthe conditionalprobabilityofapersonbeing killed when at least one beam collapses makes up 100% of indirect consequences. -The updatedt C Crep D D,/diagram in Fig. 4 can be used to assess the service life of the structure.Forexample,assumingthatthetargetvalueofrelativedirectconsequencesis 0.50,themeanvalueofservicelifeequals52.4years(RD=0.50.5=0.25)and correspondingprobabilityofhumandeath0005 . 0 ) 4 . 52 ( = s t p .Foruncorrelated resistance of beams indirect risk is RID = 1[1 - (1 - 0.0005)30] = 0.015 and belongs to the domain of tolerable risk shown in Figure 1. 6CONCLUSIONS Riskinthestructuralengineeringcanbeanalyzedandevaluatedbymeansofquantitative criteriaforidentifiedhazardscenarios,probabilitiesofconsideredeventsandtheir consequences, information obtained from inspections and assumed risk acceptance criteria. The assessment of service life of existing structures may be understood as a process of model formulation, evaluation of consequence and model updating by introducing new information. Theapproachpresented in the paper utilizes quantitative definitions of exposure, robustness andvulnerabilityofbuildingstructures.Bothdirectandindirectconsequencesassociatedwith loss of functionality, failure or collapse should be taken into consideration. The tolerable risk should be optimal with regard to benefits and costs and should be based on human safety and socio-economic values. The remaining service life of a structure may be defined in different ways, but it is strongly recommendedtoassumethesociallyallowableindirectriskrelatedtohumansafetyandto minimize direct risk connected with all types of costs. Hazardidentification,modelingofhazardscenarios,estimationofprobabilitiesand consequences compels the designer to careful examination of the whole building process and its interactionswithsafety,economy,socialandnaturalenvironment.Thepresentedapproach provides a helpful supplement to conventional assessment of building structures. REFERENCES Bertolini,L.,etal.2004.Corrosionofsteelinconcrete:prevention,diagnosis,repair.Wiley-VCH Verlag GmbH & Co. KgaA, Weinheim, ISBN 3-527-30800-8: 392. EN 1990:2002 E (2002). Eurocode - Basis of structural design. Brussels: CEN. Faber. M., et al. 2006. On the qualification of robustness of structures. Proceedings of OMAE 2006, 25th OffshoreMechanicsandArcticEngineeringConference,Hambur,Germany,June4-9,2006, (OMAEE 2006-92095). Faber,M.H.2007a.Frameworkforriskassessmentofstructuralsystems.ProceedingsofWorkshop COSTC26:Urbanhabitatconstructionsundercatastrophicevents.Prague30-31March,2007. Czech TU in Prague: 359 - 367. FaberM.,etal.2007b.Principlesofriskassessmentofengineeredsystems.10thInternational ConferenceonApplicationsofStatisticsandProbabilityinCivilEngineering.TheUniversityof Tokyo, Japan. fib Bulletin No. 34. 2006. Model Code for Service Life Design. EPFL Lausanne. ISO 2394: 1998(E). General principles on reliability for structures. Geneve: International Standard ISO. Liu, Y., Weyers, R. 1998. Modeling of the time to corrosion cracking in chloride contaminated reinforced concrete structures. ACI Journal, Vol. 95: 675 - 681. Steward,M.G.&Melchers,R.E.1997.Probabilisticriskassessmentofengineeringsystems.London: Chapman Hall. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 36Thoft - Christensen, P. 2000. Modeling of the deterioration of reinforced structures. Proc. 9th IFIP WG 7.5WorkingConferenceonReliabilityandOptimizationofStructuralSystems.TheUniversityof Michigan, Ann Arbor: 15 - 26. Wolinski,Sz.2006.Riskreliability-baseddesign.ProceedingsoftheXIth InternationalConferenceon MetalStructures:Progressinsteel,compositeandaluminiumstructures,Rzeszow,Poland,21-23 June 2006, Taylor&Francis Group: London, pp. 358 - 3599 (full text on CD- ROM). Wolinski, S., 2008. Risk-based design and assessment of concrete structures. Conference Proceeding of 6thInternationalConferenceonAnalyticalModelsandNewConceptsinConcreteandMasonry Structures. AMCM 2008. Lodz, Poland, 9-11 June, 2008, pp. 345 - 346 (full text on CD-ROM). Vrouwenvelder,T.,etal.2001.RiskAssessmentandRiskCommunicationInCivilEngineeringCIB Report: Publication 259, Rotterdam: CIB General Secretariat. Chapter 2. State-of-the-Art Report on Deterministic and Probabilistic Degradation Models Chapter 2. State-of-the-Art Report on Deterministic and Probabilistic Degradation Models Introduction 39E. Vesikari Technical Research Centre of Finland R. Landolfo University of Naples Federico II, Naples, Italy Introduction This chapter is focused on Degradations models and is organized in two parts. Part1dealswithdegradationmodesandmodelsfordifferentconstructionmaterials.The state-of-the-art related to concrete, metal, masonry and wood structures is given in four papers. The fifth paper deals with fatigue in steel and composite structures.Part2addressesapplicationofdegradationmodelstodurabilitydesign.Itconsistsoftwo state-of-the-artpapers.Thefirstpapergivesanoverviewofmoreimportantdevelopmentsand some general remarks on service life design of reinforced concrete structures with emphasis on probabilistic approach. Three levels of probabilistic methods are taken into consideration: semi-probabilistic,simplifiedprobabilisticandfullprobabilistic.Thesecondreportprovidesafirst proposalforageneralapproachtothedurabilitydesignofmetalstructuresbasedonlifetime safetyfactormethod.Allthephasesrelevanttothedesignprocedurehavebeendiscussedand defined in detail, according to a performance based approach. In modern design of structures the consideration of service life is necessary. The degradation models presented in this chapter can be used in practical service life prediction and service life designofstructures.Othersectorsofdesignwherethesemodelscanbeappliedarethe ecological and economical life cycle analyses and time variant risk analyses. Life-time structural engineering: Design for durability, life-cycle performance, maintenance and deconstruction 40E. Vesikari VTT Technical Research Centre of Finland C. Giarma Aristotle University of Thessaloniki, Thessaloniki, Greece J. Bleiziffer University of Zagreb, Faculty of Civil Engineering, Zagreb, Croatia Degradation Models of Concrete Structures 1INTRODUCTION Degradation of concrete structures is a normal consequence of the ageing process. Both physical and chemical changes occur in concrete structures with time. However, a number of parameters likethequalityofconcrete,theclimaticconditions,orthelackofmaintenance,cangreatly influence this process. Themostimportantdegradationmechanisminconcretestructuresiscor