Reliability Lessons Learned: Tragedy and Triumph

Hi,这是我刚刚无意中看到的一篇文章,看了一半,感觉很不错,,摘下来与大家分享,,大家也可点击下面的联接.:)
[url]http://www.rmspartnership.org/lessons.htm[/url]

ReliabilityLessonsLearned:

TragedyandTriumph

InR&Mengineering,justasinotherhumanendeavors,welearnfromourmistakesandoursuccesses.Theobjectivehereistoimproveoursuccessrate,asacommunityofreliabilityengineers,bysharingthehardlessonswehavelearnedinourprofession.

Someofthesetalesarehappyandsomearesad.Weoffertheminthehopethatthereissomethingtobelearnedfrombothvictoryanddefeat.Wesolicitsuggestionsfromourreaders;pleasee-mailyourcommentsandcontributionsto[email]reidwillis@juno.com[/email].Yourexperiencecouldhelpothersinthediscipline.

Topic1,ProgramPlanning

Foreveryproblemthereisasolutionthatissimple,clear,easyandwrong.

–H.L.Mencken

1.1ProgramInitiation

InaseminarontheimpactofsystemR&Mcharacteristicsonsupportrequirements,themanagerfromanaircraftmanufacturernotedthathehadfoundthat70%oftheirproducts’lifecycleproblemsarosefromtheconceptdesign.Amanagerfromanothermajormanufactureragreedthatmostoftheiroperationalsupportproblemswererootedintheinitialdesign.

Atthesameseminar,apanelfoundthatnoneofitsmembersfromgovernmentandindustryknewwhatanR&Mprogramwas,whatwasinit,orhowitwasmanagedandfunded.

Lessonlearned:

Itisessentialfortheachievementoflong-termsystemR&MperformancethataqualifiedR&MprogrammanagerbeempoweredandanR&Mprogrambeestablishedandfundedattheoutsetoftheoverallproject.

Topic2,Specifications

Becarefulwhatyouspecify.Youmightgetit.

2.1That’sObvious

Acompanydevelopedspecificationsforaliquidcoolingchamber.Thechamberwasrequiredtoacceptatemperaturesettingfromaremotecontrolcomputercontrolandmaintainit.Thetemperaturerange,rate-of-changeandtolerancewereallspecified.Thetestgrouprecommendedaddingaspecificationthatthechambermeetallrequirementswithonesetofadjustments,butmanagementsaidthatwasobvious.

Themanufacturerdeliveredthechamberanddemonstratedsatisfactoryperformance.However,differentadjustmentswererequiredtoachievespecsathigh,lowandambienttemperatures.Additionaladjustmentswereneededtodemonstratetherate-of-changerequirement.Whenitwaspointedoutthatonesetofadjustmentsshouldmeetallrequirements,themanufactureraskedformoremoneyandmoretimeonthebasisthatthiswasachangeofscope.

Lessonlearned:

Thefirstruleofwritingspecificationsistostatetheobvious.

2.2ASlightMisunderstanding

Anairlinecontractedtobuyanumberofaircraftfromthemanufacturer.Whenthefirstplanesweredeliveredtheyfailedtheairline’sacceptancetest.Theproblemturnedouttobeadifferenceininterpretationofthecontractspecs.Themanufacturerhadtobearthecostofadesignchangeandtheairline’splansfortheaircraftweredelayed.

Afterareviewthemanufacturerfoundthathistorically,mostproblemsthataroseduringacceptancetestingwereduetomisunderstandingsconcerningcontractspecifications.Theyadoptedapolicyofappointingateamofoperatorsandengineers,whohadnotbeeninvolvedintheproject,toreviewallspecificationsandacceptancetestplanstogetherwithcustomerengineers,beforethecontractwassigned.

Lessonlearned:

Allmisunderstandingsconcerningcontractspecificationsandacceptancetestprocedureswilleventuallyberesolved.Thebesttimetodothatisbeforethecontractissigned,notaftertheproductisdelivered.

2.3ObjectivesandSpecifications

Agovernmentstudyofnew-technologyprojectsfoundthatcontractualreliabilityspecificationswereoftenunrelatedtomissionreliabilityrequirements.

Lessonlearned:

Whenpreparingcontractreliabilityspecs,getyourreliabilityengineersinvolved.

Topic3,Data

Thereareonlytwokindsofreliabilityengineers:thosewhosay

“Dataistheproblem”andthosewhosay“Dataaretheproblem.”

3.1DataControl

ThecontractforprocurementofanewkindofmilitaryvehiclerequiredtheperformanceofR&Mtasksateachphaseofdesign,prototypingandproduction.Thetasksincludedreliabilityallocationandprediction,maintainabilityanalysis,failuremodeseffectsandcriticalityanalysis,andmissionsimulation.Theproducer’slogisticsmanagerwasassignedresponsibilityforsubmittinginitialR&Manalysesandperiodicupdatestothegovernment.Thecorporationgavehimhispickofnewly-hiredengineerstodotheanalysesandhischoiceofanR&Mconsultanttoshowthemhow.

Themanagerunderstoodtheimportanceofdatacontrol.Hisfirstmovewastoassignthatresponsibilitytohisstrongestengineer.TheengineerestablishedacomputerspreadsheetandpublisheditontheprojectLAN.Thespreadsheetcontainedalevel-of-indenturelisting(overallsystem,majorsubsystem,etc.)ofthevehicledesigndowntotheequipmentlevel,theequipmentMTBFandMTTRdata,andakeytothedatasource.ThesameengineerusedthespreadsheettocalculateMTBFateachlevelofindenture,forreliabilitypredictionandallocation.Thepurposesofthespreadsheetwereto:

·Publishthesystemconfigurationandequipmentfailureandrepairratestobeusedforallanalyses.

·Clearlydepictanyshortfallsintheengineeringdivisions’responsibilitiesfordatacollectionandrevision.

·NotifyallR&Manalystsofdesign,equipment,anddatachanges.

·Provideabasisforquick-responseR&Msupportoftradestudies.

·Afterasetofanalyseshadbeensubmittedforeachprocurementphase,updatetheconfigurationandequipmentdatainpreparationforthenextphase.

Whenproblemsaroseconcerninganalyticalprocedures,conclusionsandrecommendations,theR&Mmanagerandconsultantwerefreetosolvethemwithoutmajordatafumbles.AlmosthalfofthesubmittedrecommendationsresultedinR&Mdesignimprovements,whichisahighbattingaverageinthisbusiness.

ThesamekindofpositivedatacontrolwaslaterappliedtoR&Mtasksinsmallerprojectsanditprovedequallyeffectivethere.

Lessonlearned:

AttheoutsetofanR&Mprogram,establishaspreadsheetofthesystemconfiguration,equipmentfailureandrepairdata,anddatasources.Usethespreadsheetasatoolforcontrollingdatacollection,dataconsistency,andanalysisupdates.

3.2DataCollectors

Thespecificationsforanewavionicssystememphasizedreliabilityandmaintainability.Thecustomerestablishedascoringboardtoreviewflighttestdataanddeterminesystemfailurerate,repairrate,andthecauseandseverityoffailures.TheboardmembersparticipatedinplanningtestsandpreparingR&Mdatacollectionformsandinstructions.

Thefirstflighttestswereconductedbythesystemmanufacturer,usingfactoryengineerstooperatetheequipment,performmaintenance,andcollectdata.Buttheboardmembersfoundtheycouldn’tscoretheresults.Therecordsdidnotconsistentlyreportthesystemoperatinghours,thepartthatfailed,howtheequipmentwasbeingoperatedatthetimeoffailure,whatmaintenancewasperformedandhowlongittook.

Onthescoringboard’srecommendation,thecustomerprovidedobserverstooverseedatacollectionduringallsubsequentflighttestsandassociatedmaintenance.Theresultsweren’tperfectbutwhenfailuresoccurredtheboardcoulddoitsjobofoverseeingthegrowthofsystemR&Mperformance.

Lessonlearned:

Don’tdependonthemanufacturertocollecttestdatafortheanalysis.Haveyourowntrainedobserversonsitetomakesureyougetthedatayou’llneedtodoyourjob.

3.3DataAcquisitionPlanning

TheCoastGuardplannedtoputaclassofcuttersthroughamajoroverhaul.Tworeliabilityengineerswereassignedtorecommendreliabilityimprovements.Theychosemissionoperationalavailabilityasthefigureofmeritandmissionsimulationastheanalyticalapproach.

Theyforesawthatacquiringfailureandrepairdataontheships’equipmentwouldbeamajorproblem.MilitaryequipmentdatabasesdidnotapplytomostCoastGuardsystemsandtheCoastGuarddidnotkeepmaintenancehistories.Theyplannedadatastrategyinfoursteps.

1.Expendhalftheplanneddatacollectiontimeandmoneytocollectthebestdatatheycouldobtainwiththoseresources.

2.Fillintheblankswithworst-caseparametricdata.

3.RunatrialsimulationtoidentifythoseequipmentwhoseratescouldsignificantlyaffectmissionAo.

4.Expendtheremaininghalfofthedatacollectionresourcestoresolvethecriticalequipmentvalues.

Theirstrategywassuccessful.Itallowedthemtoconcentrateonthedataelementsthathadthegreatesteffectontheshipavailability.

Lessonlearned:

Plantheacquisitionofneededdataattheoutsetofthetask.Adoptastrategythatwillidentifythemostcriticaldataelementsandfocustheavailabletaskresourcesonthosevalues.

3.4DataforAgeReliabilityAnalysis

Inordertomaximizeaircraftoperationalavailability,airlinesperiodicallyreplacecertainkindsofequipmentonaschedulebeforetheyareexpectedtofail.Theykeep“rotatablepools”onhandsotheycanquicklyreplacetheequipmentandeitherdiscardtheoldunitorsenditforrefurbishmentandeventualreturntothepool.TheNavytookthesameapproachtowardsomeequipmentaboardballisticmissilesubmarines.AfterafewyearstheNavyassignedareliabilityengineertotrytheage-reliabilityproceduresthatairlinesusetorevisetheirequipmentreplacementschedulesinlightofexperience.

Age-reliabilityanalysistakesadvantageofthefactthattheschedulesarenotfollowedexactlyandequipmentsometimesfailsearlyorstaysinplacebeyonditsplannedreplacementtime.Theanalystdrawsreliabilitycurvesforeachtypeofequipment.Thelengthandshapeofthecurvetellwhethertheequipmentisbeingreplacedtoosoonortoolate.Eithercaserepresentsavoidablecosts.

Thereliabilityengineerwasabletorecommendrevisionsthatofferedsignificantsavingswithhighstatisticalcertaintyofmaintainingrequiredsystemreadinessandsafety.Thatwastheeasypartofthetask.Thedifficultpartthatrequiredmostofhistimeandexpensewaspreparingthenecessarydata.Hecomparedrefurbishmentfacilityrecords,equipmentissueandinventoryreports,andshipmaintenancerecords,andfoundthatoftentheydidnotmatch,requiringcarefulstudyandconferencewithsubmarineengineersandmaintenancesupervisorstoreconstructtheequipmentevents,operatinghoursatthetime,andcostofrepair.

Checkingwiththeairlines,hediscoveredthattheyfindfewersucherrorsbecausewhentheschedulesareinstitutedtheyincludeadatasystemtobeusedforsubsequentanalyses.Datacollectiondisciplineishighbecauseitisassociatedwithcostcenters.

Lessonlearned:

Whenpreparingaplanforscheduledpreventivereplacementorrepairofequipment,includeadatasystemtosupportlateranalysisofplaneffectivenessandopportunitiestoimprovesystemavailabilitywhilereducingmaintenancecost.

3.4AdditionalDataBenefits

WhenareliabilityengineerwasperformingR&Mpredictionsandallocationsforanewsystemtobedevelopedforagovernmentagency,henotedthathehadresearchedmuchofthesamedatabefore,inprevioustasksforthesameagency.Newdesignsoftenincludecomponentsfromexistingsystems.HeworkedwiththeagencytocreateanR&Mdatabankthatstoredlifedatafromexistingsystems,forusenotonlyinimprovingthecurrentsystemsbutalsoinpredictingtheR&Mcharacteristicsoffuturedesigns.

Lessonlearned:

Datacollectioncontributesnotonlytothefieldedproductbutalsotofutureproducts.

Topic4,Testing

TheonlycertaintyabouttestingisMurphy’sLaw.

Ifsomethingcangowrong,itwill.

Allperformancetestsareofinteresttothereliabilityengineer,notonlybecausesomeofthemareR&M-specific,butalsobecausetheycanbeincorporatedintoR&Mfailurereviewsandreliabilitygrowthanalyses.

4.1TestFacilitiesReliability

Anew-technologyairbornesystemwasbeingdevelopedtodetectenemymissiles.Themilitaryservicescreatedaspecialprojectofficetooverseetestingagainstsystemspecifications,includingR&Mspecifications.ProjectOfficereliabilityengineersparticipatedinthetestplanning.

Inanearlytesttheobjectivewastoseeifthesystemcouldreliablydetectmissileemissionsfromallangles.R&Mengineersandotherexperiencedtesterslaidthoroughplansforsystempreparationandmaintenance,testproceduresanddatacollection.Anotheragencyhadresponsibilityforproviding,authenticatingandplacingemittersthatsimulatedmissileattacks.

Thetestresultshadlimitedusefulnessbecausesomeoftheemittersfailedandtherewerenomaintenancefacilitiesorsparesonhand.

Lessonlearned:

R&Mtestplanningisthebusinessofpreparingforeveryeventuality.Onepointthatiseasytooverlookisthatwhenanotherorganizationisinvolvedtheymaynotunderstandthisprinciple.Alwaysreviewtheirplanstomakesuretheyhavealsopreparedforeveryeventuality.

4.2TestFacilitiesReliability,Revisited

Thestoryinlesson4.1hasahappyending.Inasubsequenttestthesystemwasinstalledinaircraftthatflewpastemitterssimulatingenemymissiles.Theobjectivewastoseewhetherthesystemcouldreliablydetect,pinpointandclassifythem.

ThistimetheR&Mengineersandotherplannershadlearnedtheirlesson.Theyreviewedtheplansmadebytheagencythatprovidedtheemitters,tomakesurethatemittertestcapabilitiesandspareswereonhand.Sureenough,anemitterfailedandwaspromptlyreplacedandcheckedforproperoperationwithoutinterruptingthetest.

Lessonlearned:

Seelesson4.1.

4.3MaintainabilityDemonstrators

AnArmyagencytaskedareliabilityengineertoobserveandanalyzetheresultsofamaintainabilitydemonstrationtobeconductedonaprototypeavionicssystem.Theengineerdidnothavehighhopesfortheusefulnessofthedemo.Basedonpreviousexperience,heexpectedthattheagencyconductingthedemomightnotbeabletoprovidethenecessaryskilledandtrainedpersonnel,leadingtoalast-minutedecisiontoperformthedemousingfactoryengineers.

TheArmyAviationTrainingCommandperformedthedemonstration.TheATCselectedandtrainedseveralsoldiersfromtheappropriatespecialty,includingananthropometricdistributionofaverage-size,largeandsmallmaleandfemaletechnicians.Thesoldiersperformedsystemmaintenanceactionswhilewearingutilityuniforms,flightsuitsandcoldweatherclothing.

Inadditiontomakingthenecessarystatisticalcalculationsforcomparisonagainstmaintainabilityspecifications,theobserverwasabletorecommendanthropometricdesignimprovements.Forexampleinadditiontootherproblems,thesmallestmaintainercouldnotreleaseasqueeze-typeconnectingclampandthelargestmaintainercouldnotreachintoanarrowaccessopening.

Lessonlearned:

Inamaintainabilitydemonstration,itisimportanttoconvincethecustomerthatthetestingagencymustbefurnishedadequatepersonnelandtrainingresourcesforconductingthedemo.Otherwisethereisriskofbuyingmaintenanceheadachesthatcouldhavebeenpreventedbeforeacceptingthedesign.

4.4TestAutomation

Toimproveefficiency,onecompanyautomatedtheirreliabilitytestprocess,resultinginadditionalassetsandfewerpeople.Thishadanadverseimpactonearnings,becausethecontractpaidforpeople,nottheequipment.Managementinstructedthetestunittoreduceassets.Theygotridoftheexpensiveautomationequipmentandhiredmorepeople.Notonlydidprofitsimprove,buttestqualityalsoimproved.

Lessonlearned:

Beforedecidingtoautomatethetestprocess,carefullyconsidertheimpactonfacilities,staffing,profitsandthequalityofwork.

4.5AssumedTestConditions

Acompanyrana10-daytestontheirproductinanenvironmentalchamber,usingawet-bulb/dry-bulbcontrollertomaintainandrecordhumidity.Thewetbulbanddrybulbtemperatureswerealmostexactlyequal,andwaterdropletscouldbeseenonthewindow,sotheythoughtthehumiditywasattherequiredhighpercentage.

Itturnedoutthechamberwasasdryasabone.Thewetbulbwasactuallydryandthewaterdropswerebetweenthepanesofglassinthewindow.Theyhadtorevisetheirproceduresandre-runthe10-daytest.

Lessonlearned:

Examinetestplansandprocedurescarefully.Thehiddenassumptionscanruineverything.

4.6TestFigureofMerit

Anengineerwasassignedtotakeoverthesubmissionofavailabilityreportsonanemergencycommunicationssystem.Thespecifiedrequirementwas“…availabilitynolessthan0.95…”Previousreportshadshownavailability0.99+everymonth,butthesystemwasrumoredtobeundependable.Theengineerfoundthatthemonthlyavailabilityfigurescamefromthevendor,whomeasuredsystemdowntimefromthetimethesystemwasdiscoveredtobedownuntilvendorengineersreporteditrestored:

Reporteddowntime

Availability=1––––––––––––––––––.

Calendartime

Hebegantestingatirregularintervals,andcalculatedsystemavailabilityas:

Totalsuccessfultrialtime

Availability=–––––––––––––––––––––.

Totalattemptedtrialtime

Theresultingmonthlyavailabilityaveraged0.65.VendorpaymentwassuspendeduntilthesystemmetR&Mspecifications.

Lessonlearned:

InwritingsystemR&Mspecifications,selectfiguresofmeritthathavemeaningtotheuser,anddefinehowthefiguresaretobetestedandcalculated.Readinessspecificationsforsystemsthatareseldomusedmayrequirespecialplanning.

Topic5,Analysis

Theworthofasystemanalysisdepends,morethananyotherfactor,

onhowwelltheanalystunderstandsthesystem.

–MartinBinkin

5.1ScopeoftheAnalysis

Alargesystemwasinstalledatseveralremotesites.Eachmajorsubsystemwassupportedfromadifferentcentraldepotwheresubsystempartswerebought,stored,andrepaired.Thesupportmanagertaskedareliabilityengineertolearnwhatcomputermodelswereusedoravailabletobeusedforoptimizingdepotoperationsandon-sitestocks,andrecommendchangestoimproveoverallsystemavailability.

TheengineerwasfamiliarwiththeadagefromOperationsResearchmethodologythatthefirststepshouldbetowidenthescopeoftheanalysis.Heincludedthedistributionofpartsbetweenthedepotsandthesites,andalsoexaminedthesourcesofstandardplanningfactorsthedepotmanagersusedintheircomputermodels.

Theengineerreportedthat(a)Thedepotmanagersandon-sitestockmanagersusedseveraldifferentcomputermodelsbutallwereeffectiveinoptimizingtheirlocaloperations,(b)Someplanningfactorshadbeenestablishedyearsagoas“official”butwereoutofdate,and(c)Distributionofreplacementpartsfromthedepotsandthereturnofremovedpartsfromthesiteswasunsystematic.Logisticsupportmanagerswithheldandbatchedshipmentsinawaythatwasefficientfromtheirviewpointbutdegradedoverallsystemavailability.

Lessonlearned:

Inanysystemanalysis,aninitialstepistowidenyourviewofwhatyouwereaskedtodo.Iftheclientknewwheretherealproblemwas,hewouldn’thaveneededyou.

5.2SpreadSheetMissionSimulation

TheNavywasconsideringreplacingtheenginesinaclassofships.Areliabilityengineerwasaskedtopreparemissionoperationalavailabilitycurves,tobeusedinestablishingR&Mspecificationsforreplacementenginesandsupportingtradestudies.ItwasinitiallyassumedhewouldusetheNavy’sMonteCarlo(openform)missionsimulationsoftware.However,smoothcurveswouldobviouslyberequiredandheknewthattheterm“smoothMonteCarlocurve”isanoxymoron,thereisnosuchthing.Howeverthesituationwaswellsuitedforclosed-formsimulation,usingacomputerspreadsheet.

Thespreadsheetdirectlydisplayedthefigureofmerit,inthiscasefamiliesofmissionavailabilitycurvesagainstequipmentR&Mrequirements,withoutthecharacteristicMonteCarlowobble.Inaclosed-formmodel,anychangeinthesimulatoroutputiscauseddirectlybyachangeintheinput,notbyMonteCarlorandomness.Thesmoothcurvesmadeiteasyforthecustomertoplottheengineeringoptionsandmakecost/performancecomparisons.

Inanothercaseamarineengineeringfirmhiredaconsultanttopredictmissionoperationalavailabilityofanew-technologypropulsionsystem.Themodelwastobeusedlaterincomparisons,tosupportequipmentselectionandotherdesignalternatives.Thecompanyplannedtorunitincompany-ownedMonteCarlosimulationsoftware.Theconsultantnotedthatalthoughthesystemconfigurationwasextensive,itmettherequirementsforclosedformsimulation.Hedevelopedaspreadsheetmodelthatwasmucheasiertouseandbettersuitedfortradestudies.

Lessonlearned:

InconstructinganR&Mmodelthatmaybeusedforcomparingalternatives,keepanopenmindtoclosed-formsimulation.Closed-formequationsdonothavetheflexibilityoftheMonteCarloopenform,butneitherdotheyintroducemathematicaluncertainty.

5.3Warranties

ANavyagencyassignedareliabilityengineertostudytheuseandcosteffectivenessofequipmentwarrantiesincommercialships.

Theengineerinterviewedcommercialshippersandmarineequipmentmanufacturers.Helearnedthat(a)WarrantiesforNavyequipmentwouldprobablyexpirebecausetheNavymakesadvancepurchasesandstockssparesforextendedperiods.And(b)Warrantiesareseldomenforcedbyshippersbecausetheyaregiventotheshipmaster,whohasnoincentivetoenforcethem.

Theengineerreportedonmethodsbeingusedtocalculatewarrantycosteffectiveness,andadvisedthatwarrantiesmightapplytovendorshipmentsforimmediateinstallation,butthenonlyiftheship’sofficersweregivenincentivetoenforcethem.

Lessonlearned:

Reliabilitywarrantiescannotbecosteffectiveuntilproceduresareestablishedtoensuretheirconditionsaremet,andtheofficialwhoholdsthewarrantiesisgivenincentivetoenforcethem.

5.4AnalysisFigureofMerit

Atasymposiumaprofessorpresentedapaperonriskanalysis.Theexamplewastheshipmentoffrozenfoodcontainersthatwereelectricallypoweredbytheshipper.Theanalysiswasbasedontheriskthatpowerwouldfail.

Thefirstcommentfromthefloorwasthatthefigureofmeritwasnotappropriate.Themissionoftheshipper’selectricplantwasnottoprovideuninterruptedpower,itwastomaintaincontainertemperaturesbelowaspecifiedmaximum.Itwouldbemoreappropriatetofirstdeterminethelengthoftimeacontainerwouldtoleratethelossofpowerandcontinuetomeettemperaturespecifications.Thefigureofmeritwouldbetheriskthatapowerfailureexceededtheallowedtime.

Inanothercaseamilitarysupportfacilitywantedtoimprovetheirperformance.Theyhiredaconsultingfirmtohelpthemfocusoncriticallineitems.Thefirminitiatedaprogramofmeasuringoverallperformanceastheaverageavailabilityofallsupportedsystems.Theyperiodicallycalculatedthefacility’soverallperformancefigureandidentifiedthosesystemswiththeworstavailability,foraction.

Reliabilityengineersfelttheemphasiswasbeingmisplacedonoptimizingaveragesystemavailabilityinsteadofthefacility’srealmission,whichwastosupportmilitaryreadiness.Themeasuresofoverallperformanceandsystemcriticalityshouldbeweightedbythedegreetowhicheachsystemcontributedtoreadinessofthemilitaryunitsbeingsupported.

Lessonlearned:

Investigatethoroughlywiththeclientbeforeestablishingtheobjectiveandfiguresofmeritforanalysis.Theriskisspendingtimeandresourcestoobtainresultsthatdonotsatisfyeitherofyou.

Topic6,AnalysisResults

ExperiencetellsusthatanR&Manalysisreportisunlikelygetactionforsystemimprovementiftheanalystrunsoutoftime,moneyandideasallatthesametime.

6.1EngineeringRecommendations

InarecentbrochureadvertisingR&Manalyticalsoftware,thevendorsaid,“Oncehisdatahasbeenprocessedbythe[FMECA]program,theengineercankickbackandrelax.Hisjobiscomplete.”Thisstatementwasobviouslynotwrittenbyareliabilityengineer.Infact,caseswheresystemR&Mimprovementwasobtainedbypresentingrawanalysisresultsareveryrare.Thereliabilityengineerstandsafarbetterchanceofgettingactionifheoffersspecificengineeringrecommendations.

Toshowhowthisworks,inplanningareliabilityimprovementstudyforalargeorganization,thereliabilityanalystsincludedafinalstepintheirtaskschedule,toprepareengineeringrecommendations.Duringmostoftheanalysis,theyreceivedonlylimitedcooperationfromthecorporateengineers,whowerenotaccustomedtoconsideringquantitativeR&Manalysisintheirplanning.Thereliabilityanalystspreparedadraftreportinwhichtheytranslatedtheirresultsintoalistofspecificengineeringrecommendationsforsystemredesign,equipmentselection,specialstudies,etc.

Thedraftreportgottheorganization’sengineers’fullattentionbecauseitshiftedthefocusfromwhethertheyneededtoreadthereporttohowtheyshouldrespondtoit.Theypointedout:

·Additionaldatatheanalystshadnotbeengiven,

·OtherengineeringalternativestoovercomesomeR&Mshortfalls,and

·Additionaljustificationforaction.

Withthishelp,theanalystscouldcompletethereportbyrecommendingR&Mactionsthatnowhadthesupportoftheorganization’sengineers.

Lessonlearned:

Whenschedulingananalysistask,includeafinalsteptodraftspecificengineeringrecommendations,reviewthemwithcustomerengineers,andfinalizethem.