Optimization and lifecycle engineering for design and manufacture of recycled aluminium parts

(1)

Science Arts & Métiers (SAM)

is an open access repository that collects the work of Arts et Métiers Institute of

Technology researchers and makes it freely available over the web where possible.

This is an author-deposited version published in: https://sam.ensam.eu

Handle ID: .http://hdl.handle.net/10985/11674

To cite this version :

Julien LE DUIGOU, Sverre GULBRANDSEN-DAHL, Flore VALLET, Rikard SÖDERBERG, Benoit

EYNARD, Nicolas PERRY - Optimization and lifecycle engineering for design and manufacture of

recycled aluminium parts - CIRP Annals - Manufacturing Technology - Vol. 65, p.149–152 - 2016

Any correspondence concerning this service should be sent to the repository

Administrator : archiveouverte@ensam.eu

(2)

Optimization

and

lifecycle

engineering

for

design

and

manufacture

of

recycled

aluminium

parts

Julien

Le

Duigou

a

_,

_Sverre

_{Gulbrandsen-Dahl}

b

_,

_Flore

_Vallet

a

_,

_Rikard

_So¨derberg

₍₂₎

c

_,

Benoıˆt

Eynard

a,

*

,

Nicolas

Perry

(2)

d

a_{Universite´ de}_Technologie_de_Compie`gne,_UMR₇₃₃₇_Roberval,_CS₆₀₃₁₉_–_rue_du_Dr_Schweitzer,₆₀₂₀₃_Compie`gne,_France

b_SINTEF_Raufoss_{Manufacturing}_AS,_Enggata_40,₂₈₃₀_Raufoss,_Norway

c_Chalmers_University_of_Technology,_{Maskingra¨nd}_2,₄₁₂₅₈_Go¨teborg,_Sweden

d_Arts_et_Me´tiers_ParisTech,_UMR₅₂₉₅_I2M,_Esplanade_des_Arts_et_Me´tiers,_F-33400_Talence,_France

1. Introduction

Optimized lightweight manufacturing of parts is crucial for automotiveandaeronauticalindustriesinordertostaycompetitive, andreducecostsandfuelconsumption.Hence,aluminiumbecomes anunquestionablematerialcandidateregardingthesechallenges. Nevertheless,usingonlyvirginaluminiumisnotsatisfactorysince itsextractionrequireshighuseofenergy,anditsmanufacturinghas highenvironmentalimpacts.Forthesereasons,theuseofrecycled aluminiumalloysisrecommendedsincetheirpropertiesmeetthe expected technical and environmental requirements [1]. This requires complete reengineering of the classical lifecycle of aluminium-basedproductsandseveralinterdependentdisciplines need all to be taken into account for a globalproduct/process optimization[2].Towardsthisend,thepaperproposesamethodfor sustainabilityassessmentintegrationintoproduct lifecycle engi-neering and a platform for lifecycle simulation integrating environmentalconcerns.Theplatformmaybeusedasadecision supportsysteminearly productdesignphasebysimulating the lifecycleofaproduct(frommaterialselection toproductionand recyclingphases)andcalculatingitsimpactontheenvironment.

1.1. Sustainableengineering,integratedlifecycleanddesign optimization

Design, as deﬁned in [3] is a complex and multifaceted phenomenoninvolvingatightcollaborationbetweenmulti-domain

productdesigners,amultitudeofactivitiesandprocedures,tools andknowledge,aswellasavarietyofcontexts.

Collaboration between multi-domain product designers impliesthatdifferentpointsofviewmustbetakenintoaccount toachieve thebest compromisein productdevelopment [4]. A pointofviewisthevisionandknowledgeofanexpertinvolvedina designteam[5].Anexpertmaybespecialistofaparticularlifecycle stage(e.g.manufacturing),adomain(e.g.mechanicalengineer)or cross-domain(whobringsexpertisenotlinkedtoalifestageora domain,buttoaspeciﬁcpointofviewonthewholelifecycleofa product,asforexamplethequalityengineerortheenvironmental expert).

The environmental experts often have difficulties to share environmentalinformation withotherdesign experts[6,7].This could be due to the nature of the results (e.g. environmental impacts)whicharedifficulttolinkwithotherdesignparameters (materialspecifications,geometricmodels,etc.).Itcanalsobedue totheabsenceofastandardexchangeformatthatencompasses environmentalparameters,likeSTEP(StandardfortheExchangeof Productmodeldata)thatallowsinformation exchangebetween variousexpertstools[8].Thisresultsinthelackofinteroperability between the systems used in design and those used by the environmentalexperts.

To gobeyondtheseissues, Rio [9]proposed a model-driven architecturebasedinteroperabilitymethodtoimproveexchange of information between eco-design and other design activities. Somesoftwarevendorsworkedontheintegrationofsustainability in traditional design tools like CAD systems (Solidworks from DassaultSyste`mes)ormaterialselectiontools(CESselectorfrom GrantaDesign).RussoandRizzi[10]suggestedanotherintegrated eco-design method, including shape, material and production

ARTICLE INFO Keywords: Design Optimization Lifecycle ABSTRACT

Aluminiumalloyscomponentsarenumerousinaeronauticandautomobilestructures.Despitehaving interestingmechanicalpropertiesforlightweightsolutions,theextractionofvirginaluminiumstillhas negativeimpactsontheenvironment.Asolutionistouseanincreasedrateofrecycledaluminiumin structuralparts.Thisrequiresaglobaloptimizationofthepartdesignandmanufacture.Theproposed workdetailstheadvancedoptimization techniquesusedforproductandprocessdesignintegrating environmentalconcerns.Themethodologyisimplementedandtestedonanindustrialcasethatresultsin arecyclingrateof75%inhigh-endstructuralcomponentbasedonwroughtaluminiumalloys.

ß2016CIRP.

*Correspondingauthor.

E-mailaddress:benoit.eynard@utc.fr(B.Eynard).

ContentslistsavailableatScienceDirect

CIRP

Annals

-

Manufacturing

Technology

j o u r n a lh o m e p a g e :h t t p : / / e e s . e l s e v i e r . c o m / c i r p / d e f a u l t . a s p

http://dx.doi.org/10.1016/j.cirp.2016.04.111

(3)

assessmentintegratingLifeCycleAssessment(LCA)withCADand Finite Element Analysis.But those modules focus on a speciﬁc designstageanddo notconsiderenvironmentalimpactson the wholelifecycle.Otherresearchers,likeDufreneetal.[11]proposed an integrated eco-design methodology that improves both environmentalimpactsandtechnicalcharacteristics.

Oneofthemajordifﬁcultiestheenvironmentalexpertshaveto copewithisthelackofinformationespeciallywhentheytryto performLCA [12,13]. LCA is time and resource consuming and requiresahugeamountofheterogeneousdatafromalloverthe extended enterprise. Stark and Pfo¨rtner [14] proposed an ontologicalapproachtoassesssustainabilitybasedoninformation fromITsystemsandcalculationrules.Someof thisinformation could be extracted from the digital mock-up. This requires integration between CAD, Product Lifecycle Management and LCA,but to make a clearand usefulanalysis of environmental impacts,thisinformationmustbespecializedandaccurate[15]. 1.2. Objective

Adesignmethodologythatintegrateswholelifecycle environ-mentalimpactassessmentintoaproductdesignoptimizationloop is not yet realized. In order to include the entire lifecycle environmental impact,theassessment ofthe optimized design should take into consideration the extraction, manufacturing, distribution,useandendoflife.Thisleadstoamethodologythat integrates environmental concerns into a closed loop design optimization.Theoptimizationincludesmaterialchoice(basedon recycledaluminiumcompositions),topologyoptimizationforthis particularmaterial,optimizationofprocessesandtolerances,and simpliﬁedenvironmentalassessment.Theobjectiveistopropose newmaterials,processesand parts design that fulﬁlhighlevel requirementsanddecreasetotalenvironmentalimpacts.

2. Proposal:theSuPLightclosedloopdesignmethodology ThemethodologyisdevelopedintheEuropeanprojectSuPLight (Sustainableand efﬁcientProductionof Lightweightsolutions). SuPLightisamultidisciplinaryresearchproject,combiningphysics at the atomic scale level, metallurgy, continuum mechanics, structuralmechanics,optimizationalgorithms,toleranceanalysis, andmanufacturingandlifecycleassessment.SuPLightproposesto reduceweightinstructural partsand improvetheholistic eco-designofaluminiumwroughtalloysandtobuildnovelsustainable industrymodelswithaholisticlifecycleapproach.

Thismethodologyisbasedonanoptimizationlooponmaterial, parttopology,processesandpartstolerancesandenvironmental assessment(Fig.1).Itissupportedbyanintegratedoptimization platform that supports automatic data exchange between the differenttools.

Thematerial phase determinesthematerial andmechanical properties of the alloys which are later used by the design optimizationandprocessphasestocalculatethebehaviourofthe alloyparts.

Thedesignoptimizationperformsananalysisinordertofind alternativetopologiesforthepartthatfulfilthestiffness,weight criteria, eigenvalues and centre of gravity requirements. The outputofthisphaseincludesthegeometrydefinitionusedasan inputtotheprocessoptimizationphase.

Theprocessoptimizationphaseassessestheforgingprocessfor thepartdeﬁnedbythedesign optimization.Theoutputofthis phase includesthegeometricdeﬁnitionfor thetolerancephase input.

The tolerance optimizationusesa meta-modelfor tolerance analysis and geometric variation simulation to provide rapid resultsbasedonamoreextensivecomputationmodule.

The environmental assessment computes environmental impactsbasedonthecharacteristicsoftheproduct.Thesimpliﬁed lifecycleassessmentofthepart(initscontextofusage)isbasedon theinputsfromtheotherphases.

2.1. Material

Basedonthechemicalcompositionofthetargetalloyandits processingpaththeanalysisreturnsthesetofrelevantmaterial properties.Thecomposition ofthealloycanbespeciﬁedby its unique IDbased on thepredeﬁnedset of alloysor by element composition.Theprocessingpathcanbeeither(1)diecastand forging, (2) continuous casting and forging, or (3) continuous casting,extrusionandforging.

The material properties are calculated using two different scenarios:basicusageandadvanced usage.Inthecaseof basic usage,thematerialpropertiesareevaluatedbymeansof meta-modelsbasedon theexperimental data.Ontheotherhandthe advancedoperationsallowtheevaluationofmechanical proper-tiesofthealloybymetallurgicalprecipitationhardeningmodel.In the initial stage the model is based on classical Kampmann– Wagnermodel[16].Duetotheknownlimitationsofthismodel withrespecttothermodynamics,theadditionalmodelSFFKwas implemented [17]. The latter model gives the possibility to overcomelimitationsofKampmann–Wagnermodelapproachby allowing modelling of multicomponent and even multiphase systems.Thedevelopedframeworkallowswiderangeof adapta-tionsofthemodelandintroductionsofadditionalcomponents.As aﬁrstvalidation,successfulinitialtestingwasperformedusingthe thermodynamicdatafromliteratureandothersources.

2.2. Topologyoptimization

In structural optimization, the use of differentsets of data representingamathematicalmodeldescribesthebehaviourofa structure.Differentcontrolparametersaretunedbyasetofdesign variablestoﬁndasituationinwhichthestructuremeetsagiven property.Shapeoptimizationconsistsofoptimizingthestructure bychangingtheshape.Shapeoptimizationhasaninterdisciplinary character,meaningitcanbeusedonawiderangeofproblems.This kind of problems involves mathematical disciplines as partial differential equations, approximations of these and theory of nonlinear mathematical programming.This processis an auto-matedandintegratedtaskintheproposedframework.

2.3. Process/manufactureengineering

Theuseof specializedcodes topredictmaterialﬂow during forginghasbeendevelopedtogetherwithoftherapidgrowingof computer-aided engineering, e.g. FORGE NxT. In contrast to structural design,thecodes forhandlingforgingare typicalset up to manage large strains and deformations. Combining the predictivecapabilitiesinsuchcodeswithanoptimizationengine givesanadvantagefordesigningsustainableandoptimalforging

Fig.1.SuPLightoverallmethodology.

J.LeDuigouetal./CIRPAnnals-ManufacturingTechnology65(2016)149–152

(4)

processes.Theaimoftheforgingphaseistosortoutthebestdesign whichwillﬁllthedieswithoutﬂowerrors.

Oneapproachtoreachthisgoalistouseapre-shapedesignand lettheoptimizationenginescaletheshapebasedonthecentreof gravity. A second approach is to utilize the same principals describedinthestructuraloptimizationonthepre-shape.Inthis workit was chosen todo thescaling approach on the forging optimization,giventhatthestructuralstrategyhasalreadybeen demonstratedontheproductdesign.

Thisphaseisdependentonreceivingtwocentralinputs.Firstit needsageometrydeﬁnitionofthedesiredendproduct;seconda description of the material to use. The end product is then subjectedtoa‘‘forgeability’’assessmentwhichsortsouttosmooth thepartincludingdraftanglesandradiuses.Thenextstepisto subtractthediesincludingthenecessaryﬂeshthickness.Thedies arethenindividuallypreparedforthepre-forgingsetup.

Inthepre-forgingsetup,thepreparedgeometriesofthedies are converted to a representative numerical node mesh. To complete the set-up process, all process specific details are decided. Based on the material composition a forging billet temperatureisset,togetherwiththeinteractioncoefficientsand heattransfernumber.Furthermore,thetypeofpressandstroke kinematicsisdefined,togetherwiththestoringincrementdensity. Theprojectisthenreadyfortheprocessoptimizationphase.

To eliminate designs that could represent an under filling, constrains are introduced in the optimization. Two sets of constrainsareused.First thebilletneedstobeincontact with allsurfacesthatdescribetheendproduct.Lackofsuchcontactwill breach the constraint and the result is returned as not valid. Second,thescalarthatrepresentsanevaluationofflowerroror self-contactwithinthedomainofthedesiredendproductneedsto be zero. Results which deviate from zero are breaching the constraintandarethusnotvalid.Theprocessoptimizationphaseis setuptodisregardthenotvaliddesignandusesacostfunctionto findthelowestweightoftheinitialpre-shapethatstillupholdsthe constraints.

2.4. Tolerancesdefinition

Allmanufacturingprocessesaresubject tovariations,which may affect the way that the ﬁnal product will meet its speciﬁcations.Variationsourcescanbeinmaterialpropertiesor manufacturing conditions such as stamping pressure, thermal conduction, temperature,friction,etc. It is thereforeof highest importance to consider this variation in the input parameters duringdesignphasesandtrytopredicthowtheywillaffectthe output performance parameters. In assembled products, the geometricdeformationofindividualpartsis veryimportantfor of manufacturing operations and for variation simulation of assemblies.

Topredictvariationinthefinalgeometryduetovariationin material and manufacturing parameters, variation simulation based on statistical calculations and expected distributions for input parameters is used. Since variation simulation will be conductedoncomputationallyexpensivefiniteelementsmodels, metamodelsincombinationwithMonteCarlosimulationisused topredictthefinalgeometricvariation[18].

2.5. Environmentalassessment

Thefinalaimofthisphaseistoprovideanevaluationofthe designedpartwholelifecycleenvironmentalimpact.This evalua-tion realized by an LCA tool is coupled with an eco-design approach.Thedevelopedapproachconsistsatfirstoflinkingthe key environmental indicators evaluatedby the LCAtool to the different parameters in the whole product lifecycle, such as material,typeofmanufacturingprocess,etc.Thenitlinks these parameters to possible improvements and solutions to reduce environmentalimpact[19].Thefinaloutputoftheenvironmental assessmentistheimpactoftheactualdesignandimprovement

recommendations.Theproductdesignerschoosetovalidateornot someoftheserecommendationsandrestarttheentire optimiza-tionloopuntiltheenvironmentalimpactissatisfactory.

3. Casestudy

3.1. Testcasedefinition

Theconsideredpartforthecasestudypresentedinthissection, is a front lowercontrolarm,which is a partofthesuspension systemofacar(seeFig.2).Thechosencontrolarmismanufactured fromAA6082whichisahighstrengthwroughtaluminiumalloy. Thebaselineprocessforthemanufacturingofthecontrolarmis: (1)manufacturingofAA6082ingotsfromvirginaluminium and treatedproductionscrap;(2)extrusionofingotstoproducethe AA6082rods;(3)cuttingrodsintopiecesandannealingtosoften thebilletandmakeiteasiertoforge;(4)rodsforging;(5)ageing rods;(6)machiningandassemblyofthecontrolarm.

Theproduction(orpre-consumer)scrapistreatedtoremove theﬂuidsanddirtandisthenreusedinthemanufacturingofthe AA6082 ingot.The aluminiumrecycling process(postcustomer scrap)isdescribedin[20].

Threescenarioswereproposed,withthepurposeofspecifying theoptimalaluminiumalloypropertiestobeusedforthecontrol armtoachievethebestglobalperformance(intermsofresistance, rateofrecycledaluminiumorenvironmentalimpacts).

DetailsofthesimpliﬁedLCAmodelforthepresentpartdesign areavailablein[19].The3scenarioswhicharecomparedtothe presentdesign andtechnology are:(1)newalloybasedon 75% post-consumer scrap; (2) optimized product and process with respecttoweight;(3)newalloybasedon75%post-consumerscrap andoptimizedproductandprocesswithrespecttoweight.They usetheprimaryaluminiumandrecycledaluminiumcompositions presentedinTable1.

3.2. Platformimplementation

The different phases are implemented by a set of plugins: material plugin, topology optimization plugin, process plugin, tolerances plugin and environmental plugin. The plugins have their own interfaces and some of them may also be used as independent software. In this case, they have two functioning modes(independentandintegratedmode).Acommonformatwas speciﬁed toseta standardized formatfor enabling information exchange and management based on a dictionary. All ‘‘Simple ObjectAccessProtocol’’messagessentbetweenthe communicat-ingmodulesfollowtherulessettledbythestandardizedformat.In theintegratedmode,theyrunaspartoftheclosed-loopandthey areprovidedwithacommoninterface,accessibleonlinewithinthe SuPLight simulation platform (https://collab.suplight.eu/sim/). Web services were deﬁned and implemented for each plugin,

Fig.2.Thefrontlowercontrolarm.

Table1

Aluminium(AL)alloyscompositions.

SI Mg Fe Cu Mn Zn

PrimaryAL 0.98 0.69 0.15 0 0.43 0.03 RecycledAL 0.92 0.75 0.16 0.01 0.4 1.1

(5)

accordingtotheirlegacyformat.Servletshandlewebservicecalls fromtheSuPLightplatformandtranslatesthemintolegacyformat. Themoduleforinformationmanagement(ModeFrontier)handles theexecutionofthepluginsintheoptimizationloop.Atthislevel, the plugin sequence and the mapping between the input and outputofeachpluginaremanaged.Inthefront-end,theexecution sequencemaybesetfromtheGraphical UserInterfaceand the executionofthesimulationsequencemaybevisualized step-by-step(resultsfromeachplugin)orat-once(ﬁnalresult).

4. Resultsanddiscussion

When the optimal parameterization strategy was chosen, automaticoptimizationinModeFrontiergeneratedweightsavings of6.26%whichisaquitegoodresultconsideringtheseproducts havebeeninmassproductionforyears.

ThereferencemassnecessarytoforgethereferenceFLCAhas been taken from the production input and is measured to be 2700g.After40runswithintheoptimizationpluginthelowest initialpre-shape weightisestimated tobea reduction ofmass comparedtothereferencemassof16.4%.Asecondaryeffectwith the new reduced mass is a lowering of forging loads from 1650tonnesto1600tonnes.

TheoptimizationresultsaresummarizedinTable2.

Thereductionofweightinﬂuencesthewholelifecycleofthe productandreduces thetotalenvironmental impact.Incase of FLCA,reductionofweightcombinedwithusingrecycled alumini-umand changingthe productionrouteresult ina reduction of environmentalimpacts between24 and50% atthehighest(for water withdrawal – Scenario 2), see Fig. 3. The results from simpliﬁedLCAintheSuPLightframeworkarepresentedinFig.3as fractions of environmental impacts of the present design and technology.

5. Conclusionandfuturework

Thispaperproposedaframeworkforanoptimizationloopof structural part design including environmental assessment throughanaluminium partredesign integratinga highlevelof post-customer scrap. This framework is implemented on a numericalplatformandtestonanautomotivepart.

Thisworkshowsthatanintegratedapproachfrommaterialto production and environmental assessment is necessary. The needed information is processed through the different phases and several optimization loops are necessary to ﬁnd the best solution.

Otherlifecyclephasesshouldbeintegratedtoproposeamore completeanalysis.Theproductionphaseisreducedtotheforging stepinthepresentedprocess.Amorecomplexoptimizationofthe productionroutescanoffernewperspectives.

Thisalsoofferstheperspectiveforthedevelopmentofaspecific reverselogisticbusinessthatselectsandsortsendcustomerscrap dependingonthealuminiumcomponentforspecificapplications. Thisnewindustrialmodelcouldallowtheimplementationofthe proposedmethodologywithsufficientrecycledaluminium quan-tityandanacceptablechemicalcomposition.

Acknowledgements

TheauthorswouldliketoacknowledgetheEuropean Commis-sion for its ﬁnancial support throughthe SuPLight FP7project (grantagreementn8263302)andsupportfromtheresearchcentre SFIManufacturingsupportedbytheResearchCouncilofNorway.

References

[1]DuﬂouJR,TekkayaAE,HaaseM,WeloT,VanmeenselK,KellensK,DewulfW, Paraskevas D (2015) Environmental Assessmentof SolidState Recycling RoutesforAluminiumAlloys:CanSolidStateProcessesSigniﬁcantlyReduce theEnvironmentalImpactofAluminiumRecycling?CIRPAnnals– Manufactur-ingTechnology64(1):37–40.

[2]PerryN,BernardA,Bosch-MauchandM,LeDuigouJ,XuY(2014)Proposalofan ArchitecturalSolutionforEconomicandEnvironmentGlobalEco-cost Assess-ment:ModelsCombinationAnalysis.inHenriquesE,Pec¸asP,SilvaA,(Eds.)

TechnologyandManufacturingProcessSelection:TheProductLifeCycle Perspec-tive,239–256.

[3]Blessing LT,ChakrabartiA(2009) DRM–ADesign ResearchMethodology,

Springer-Verlag,London.

[4]Sohlenius G(1992)Concurrent Engineering.CIRP Annals –Manufacturing Technology41(2):645–655.

[5]BrissaudD,TichkiewitchS(2001)ProductModelsforLifeCycle.CIRPAnnals– ManufacturingTechnology50(1):105–108.

[6]LindhalM(2006)EngineeringDesigners’ExperienceofDesignfor Environ-mentMethodsandTools-RequirementDeﬁnitionsFromanInterviewStudy.

JournalofCleanerProduction14(5):487–496.

[7]KozemjakindaSilvaM,RemyS,ReyesT(2015)OnProvidingDesignProcess Information to theEnvironmental Expert. Research in EngineeringDesign

26(4):327–336.

[8]ChandrasegaranSK,RamaniK,SriramRD,Horva´thI,BernardA,HarikRF (2013)TheEvolution,Challenges,andFutureofKnowledgeRepresentationin ProductDesignSystems.Computer-AidedDesign45(2):204–228.

[9]RioM,ReyesT,RoucoulesL(2014)FESTivE:AnInformationSystemMethodto Improve Product Designers and Environmental Experts Information Exchanges.JournalofCleanerProduction83:329–340.

[10]RussoD,RizziC(2014)StructuralOptimizationStrategiestoDesignGreen Products.ComputersinIndustry65(3):470–479.

[11]DufreneM,ZwolinskiP,BrissaudD(2013)AnEngineeringPlatformtoSupport aPracticalIntegratedEco-DesignMethodology.CIRPAnnals–Manufacturing Technology62(1):131–134.

[12]CurranMA(2006)LifeCycleAssessment:PrinciplesandPractice.EPA/600/R-06/ 060.

[13]ISO14040(2006)EnvironmentalManagement–LifeCycleAssessment– Princi-plesandFramework,InternationalStandardOrganisation,Geneva.

[14]StarkR,Pfo¨rtnerA(2015)IntegratingOntologyintoPLM-ToolstoImprove SustainableProductDevelopment.CIRPAnnals–ManufacturingTechnology

64(1):157–160.

[15]Pavkovic´ N,SˇtorgaM,Bojcˇetic´ N,Marjanovic´ D(2013)FacilitatingDesign CommunicationThroughEngineeringInformationTraceability.Artificial Intel-ligenceforEngineeringDesignAnalysisandManufacturing27(2):105–119.

[16]MyhrOR,GrongØ.,PedersenKO(2010)ACombinedPrecipitation,Yield Strength,andWorkHardeningModelforAl-Mg-SiAlloys.Metallurgicaland MaterialsTransactionsA41(9):2276–2289.

[17]SvobodaJ,FischerFD,FratzlP,KozeschnikE(2004)ModellingofKineticsin Multi-ComponentMulti-PhaseSystemsWithSphericalPrecipitates:I:Theory.

MaterialsScienceandEngineeringA385(1–2):166–174.

[18]HoffensonS,DagmanA,So¨derbergR(2015)ToleranceOptimisation Consid-eringEconomicandEnvironmentalSustainability.JournalofEngineering De-sign25(10–12):367–390.

[19]AndriankajaH,ValletF,LeDuigouJ,EynardB(2015)AMethodtoEcodesign StructuralPartsintheTransportSectorBasedonProductLifeCycle Manage-ment.JournalofCleanerProduction94:165–176.

[20]GlobalAluminiumRecycling(2009)ACornerstoneofSustainableDevelopment,

InternationalAluminiumInstitute.

Table2

Optimizationresultsfromthetestcase.

Recyclinglevel Shapeoptimization Processoptimization Result 75% 93.74%weight 82.67%weight

Fig.3.SimpliﬁedLCAresultsfromtestcase.

J.LeDuigouetal./CIRPAnnals-ManufacturingTechnology65(2016)149–152

152

Powered by TCPDF (www.tcpdf.org) Powered by TCPDF (www.tcpdf.org) Powered by TCPDF (www.tcpdf.org) Powered by TCPDF (www.tcpdf.org)