六西格瑪應(yīng)用質(zhì)量體系手冊議程

1、 QSM 754SIX SIGMA APPLICATIONS AGENDADay1AgendaWelcome andIntroductionsCourseStructureMeeting Guidelines/Course Agenda/ReportOut CriteriaGroupExpectationsIntroductiontoSix Sigma ApplicationsRedBeadExperimentIntroductiontoProbabilityDistributionsCommonProbabilityDistributionsandTheirUsesCorrelationAn

2、alysisDay2AgendaTeam ReportOuts on Day1 MaterialCentral Limit TheoremProcess CapabilityMulti-Vari AnalysisSampleSizeConsiderationsDay3AgendaTeam ReportOuts on Day2 MaterialConfidence IntervalsControl ChartsHypothesis TestingANOVA(AnalysisofVariation)ContingencyTablesDay4AgendaTeam ReportOuts on Prac

3、ticumApplicationDesignofExperimentsWrap Up -Positives andDeltasClassGuidelinesQ&AaswegoBreaksHourlyHomeworkReadingsAsassignedinSyllabusGradingClassPreparation30%Team ClassroomExercises30%Team Presentations40%10MinuteDailyPresentation (Day2and3)onApplicationofpreviousdayswork20minutefinalPracticum ap

4、plication(Lastday)Copy on Floppyaswell as hardcopyPowerpoint preferredRotatePresentersQ&AfromtheclassINTRODUCTIONTOSIX SIGMA APPLICATIONSLearningObjectivesHave abroadunderstandingofstatisticalconceptsandtools.Understand howstatisticalconceptscanbeusedtoimprove business processes.Understand therelati

5、onshipbetweenthecurriculumandthe fourstepsixsigmaproblemsolving process(Measure, Analyze, Improveand Control).What is SixSigma?A PhilosophyA QualityLevelA StructuredProblem-Solving ApproachA ProgramCustomer Critical To Quality (CTQ) CriteriaBreakthrough ImprovementsFact-driven, Measurement-based, St

6、atistically Analyzed PrioritizationControlling the Input & Process Variations Yields a Predictable Product6s = 3.4 Defects per Million OpportunitiesPhased Project:Measure, Analyze, Improve, ControlDedicated, Trained BlackBeltsPrioritized ProjectsTeams - Process Participants & OwnersPOSITIONINGSIXSIG

7、MATHEFRUITOFSIX SIGMAGround FruitLogic and IntuitionLow Hanging FruitSeven Basic ToolsBulk of FruitProcess Characterization and OptimizationProcess EntitlementSweet Fruit Design for ManufacturabilityUNLOCKINGTHE HIDDENFACTORYVALUE STREAM TO THE CUSTOMERPROCESSES WHICH PROVIDE PRODUCT VALUE IN THE CU

8、STOMERS EYESFEATURES OR CHARACTERISTICS THE CUSTOMER WOULD PAY FOR.WASTE DUE TO INCAPABLE PROCESSESWASTE SCATTERED THROUGHOUT THE VALUE STREAMEXCESS INVENTORYREWORKWAIT TIMEEXCESS HANDLINGEXCESS TRAVEL DISTANCESTEST AND INSPECTIONWaste is a significant cost driver and has a major impact on the botto

9、m line.CommonSix Sigma ProjectAreasManufacturing DefectReductionCycleTime ReductionCost ReductionInventoryReductionProduct Development andIntroductionLaborReductionIncreasedUtilizationofResourcesProduct Sales ImprovementCapacityImprovementsDeliveryImprovementsTheFocusofSix Sigma.Y = f(x)All critical

10、 characteristics (Y) are driven by factors (x) which are “upstream” from the results.Attempting to manage results (Y) only causes increased costs due to rework, test and inspectionUnderstanding and controlling the causative factors (x) is the real key to high quality at low cost.INSPECTION EXERCISET

11、henecessity of training farmhandsfor first class farms in thefatherlyhandlingoffarm livestockisforemostinthemindsoffarmowners.Sincetheforefathersofthe farmowners trainedthe farmhandsfor first class farms in thefatherlyhandlingoffarm livestock,the farmowners feeltheyshouldcarryonwiththefamily traditi

12、onoftrainingfarm hands of first class farms in thefatherlyhandlingoffarm livestockbecause theybelieveitisthebasisofgoodfundamentalfarm management.Howmanyfscan youidentifyin1 minuteofinspection.INSPECTION EXERCISEThenecessity of*trainingf*armhandsf*or f*irstclassf*armsinthe f*atherlyhandlingof*f*arml

13、ivestock is f*oremostinthemindsof* f*arm owners.Since thef*oref*athers of*thef*armowners trainedthe f*arm hands f*orf*irst class f*armsinthef*atherly handling of*f*armlivestock, thef*armownersf*eeltheyshouldcarryonwiththef*amilytraditionof* training f*arm hands of*f*irstclassf*arms in thef*atherlyha

14、ndlingof* f*arm livestockbecause theybelieveitisthebasisof* goodf*undamental f*arm management.Howmanyfscan youidentifyin1 minuteofinspection.36totalareavailable.SIXSIGMACOMPARISONSix Sigma Traditional “SIX SIGMA TAKES US FROM FIXING PRODUCTS SO THEY ARE EXCELLENT, TO FIXING PROCESSES SO THEY PRODUCE

15、 EXCELLENT PRODUCTS” Dr. George Sarney, President, Siebe Control SystemsIMPROVEMENTROADMAPBreakthroughStrategyCharacterizationPhase 1:MeasurementPhase 2:AnalysisOptimizationPhase 3:ImprovementPhase 4:ControlDefine the problem and verify the primary and secondary measurement systems.Identify the few

16、factors which are directly influencing the problem.Determine values for the few contributing factors which resolve the problem.Determine long term control measures which will ensure that the contributing factors remain controlled.ObjectiveMeasurementsare critical.Ifwecantaccurately measuresomething,

17、wereallydontknowmuch about it.Ifwedontknow muchaboutit, we cant controlit.Ifwecantcontrol it,weareatthe mercy of chance.WHYSTATISTICS?THEROLEOFSTATISTICS IN SIXSIGMA.WEDONTKNOW WHATWEDONTKNOWIFWEDONTHAVE DATA, WE DONT KNOWIFWEDONTKNOW,WECANNOT ACTIFWECANNOT ACT,THE RISKISHIGHIFWEDOKNOW ANDACT, THERI

18、SK IS MANAGEDIFWEDOKNOW ANDDONOTACT,WEDESERVE THELOSS.DR.MikelJ.HarryTOGETDATAWEMUST MEASUREDATA MUSTBECONVERTED TO INFORMATIONINFORMATIONISDERIVED FROMDATATHROUGH STATISTICSWHYSTATISTICS?THEROLEOFSTATISTICS IN SIXSIGMA.Ignoranceisnot bliss,itisthefoodoffailure andthebreedingground forloss.DR.MikelJ

19、.HarryYearsagoastatisticianmighthaveclaimed thatstatisticsdealtwith theprocessing of data.Todays statisticianwill be morelikely to saythat statisticsisconcerned withdecisionmakinginthe faceofuncertainty.BartlettSalesReceiptsOnTime DeliveryProcess CapacityOrderFulfillmentTimeReductionofWasteProduct D

20、evelopment TimeProcess YieldsScrapReductionInventoryReductionFloorSpaceUtilizationWHAT DOESITMEAN?RandomChance or Certainty.Whichwouldyouchoose.?LearningObjectivesHave abroadunderstandingofstatisticalconceptsandtools.Understand howstatisticalconceptscanbeusedtoimprove business processes.Understand t

21、herelationshipbetweenthecurriculumandthe fourstepsixsigmaproblemsolving process(Measure, Analyze, Improveand Control).REDBEADEXPERIMENTLearningObjectivesHave an understandingofthe differencebetweenrandomvariation anda statisticallysignificantevent.Understand thedifference betweenattemptingtomanagean

22、outcome(Y)asopposedtomanagingupstreameffects (xs).Understand howtheconceptofstatisticalsignificancecan be usedtoimprovebusinessprocesses.WELCOME TO THEWHITEBEAD FACTORYHIRINGNEEDSBEADSAREOUR BUSINESSPRODUCTION SUPERVISOR4 PRODUCTIONWORKERS2 INSPECTORS1 INSPECTIONSUPERVISOR1 TALLY KEEPERSTANDINGORDER

23、SFollowthe processexactly.Donotimprovise or varyfromthedocumentedprocess.Your performance willbebasedsolely on yourabilitytoproduce white beads.Noquestionswillbeallowed after theinitial training period.Your defectquotaisnomore than5offcolorbeadsallowedperpaddle.WHITEBEAD MANUFACTURINGPROCESSPROCEDUR

24、ESTheoperatorwilltake thebead paddleintherighthand.Insertthe beadpaddle at a45degree angle intothe beadbowl.Agitate thebead paddlegentlyinthe beadbowluntilallspaces arefilled.Gentlywithdrawthe beadpaddle fromthe bowlata45degreeangleand allow thefree beads torunoff.Without touching thebeads,showthepa

25、ddle to inspector#1andwaituntiltheoff color beads aretallied.Move to inspector#2andwaituntiltheoff color beads aretallied.Inspector#1and #2 showtheirtalliestotheinspectionsupervisor.Iftheyagree,the inspectionsupervisorannouncesthe count andthetallykeeper willrecord theresult.Ifthey do notagree,the i

26、nspectionsupervisorwill directtheinspectorstorecount thepaddle.When thecountiscomplete,the operator willreturn allthebeadstothebowl andhand thepaddletothe nextoperator.INCENTIVEPROGRAMLowbeadcounts willberewardedwitha bonus.High beadcounts willbepunishedwitha reprimand.Your performance willbebasedso

27、lely on yourabilitytoproduce white beads.Your defectquotaisnomore than7offcolorbeadsallowedperpaddle.PLANTRESTRUCTUREDefectcounts remaintoohighforthe plant to be profitable.Thetwo bestperformingproduction workerswillberetainedandthe twoworstperforming productionworkerswill be laidoff.Your performanc

28、e willbebasedsolely on yourabilitytoproduce white beads.Your defectquotaisnomore than10off color beads allowedper paddle.OBSERVATIONS.WHAT OBSERVATIONSDIDYOU MAKEABOUTTHISPROCESS.?TheFocusofSix Sigma.Y =f(x)Allcriticalcharacteristics (Y)aredriven by factors(x) which are“downstream”fromtheresults.Att

29、empting to manageresults (Y)only causesincreasedcostsdue to rework,testandinspectionUnderstanding andcontrollingthecausative factors(x) is thereal keytohigh qualityatlow cost.LearningObjectivesHave an understandingofthe differencebetweenrandomvariation anda statisticallysignificantevent.Understand t

30、hedifference betweenattemptingtomanageanoutcome(Y)asopposedtomanagingupstreameffects (xs).Understand howtheconceptofstatisticalsignificancecan be usedtoimprovebusinessprocesses.INTRODUCTIONTOPROBABILITYDISTRIBUTIONSLearningObjectivesHave abroadunderstandingofwhat probability distributionsare andwhyt

31、heyareimportant.Understand therole thatprobabilitydistributionsplay in determining whetheraneventisarandomoccurrenceorsignificantly different.Understand thecommonmeasuresusedtocharacterizeapopulation centraltendencyand dispersion.Understand theconcept of Shift &Drift.Understand theconcept of signifi

32、cancetesting.WhydoweCare?Anunderstanding of Probability Distributionsisnecessary to:Understand theconcept anduseofstatisticaltools.Understand thesignificanceofrandom variationineverydaymeasures.Understand theimpactofsignificance on thesuccessful resolutionofaproject.IMPROVEMENTROADMAPUses of Probabi

33、lity DistributionsBreakthroughStrategyCharacterizationPhase 1:MeasurementPhase 2:AnalysisOptimizationPhase 3:ImprovementPhase 4:ControlEstablishbaselinedatacharacteristics.Project UsesIdentifyandisolatesources of variation.Usethe conceptofshift&drifttoestablishprojectexpectations.Demonstratebeforean

34、d after resultsare notrandomchance.Focus on understanding the conceptsVisualize the conceptDont get lost in the math.KEYS TO SUCCESSMeasurementsare critical.Ifwecantaccurately measuresomething,wereallydontknowmuch about it.Ifwedontknow muchaboutit, we cant controlit.Ifwecantcontrol it,weareatthe mer

35、cy of chance.TypesofMeasuresMeasureswherethemetric is composed of aclassificationinone of two(ormore)categoriesiscalledAttributedata.This dataisusuallypresentedasa“count” or “percent”.Good/BadYes/NoHit/Missetc.Measureswherethemetric consists of anumber which indicatesa precisevalueiscalledVariableda

36、ta.TimeMiles/HrCOIN TOSSEXAMPLETake acoinfrom yourpocket andtoss it 200times.Keep track of thenumberoftimesthe coinfallsas“heads”.When complete,theinstructorwill askyoufor your“head” count.COIN TOSSEXAMPLE1301201101009080701000050000Cumulative FrequencyResults from 10,000 people doing a coin toss 20

37、0 times.Cumulative Count1301201101009080706005004003002001000Head CountFrequencyResults from 10,000 people doing a coin toss 200 times.Count Frequency130120110100908070100500Head CountCumulative PercentResults from 10,000 people doing a coin toss 200 times.Cumulative PercentCumulative FrequencyCumul

38、ative PercentCumulative count is simply the total frequency count accumulated as you move from left to right until we account for the total population of 10,000 people.Since we know how many people were in this population (ie 10,000), we can divide each of the cumulative counts by 10,000 to give us

39、a curve with the cumulative percent of population.COIN TOSSPROBABILITYEXAMPLE130120110100908070100500Cumulative PercentResults from 10,000 people doing a coin toss 200 timesCumulative PercentThis means that we can now predict the change that certain values can occur based on these percentages.Note h

40、ere that 50% of the values are less than our expected value of 100.This means that in a future experiment set up the same way, we would expect 50% of the values to be less than 100. COIN TOSSEXAMPLE1301201101009080706005004003002001000Head CountFrequencyResults from 10,000 people doing a coin toss 2

41、00 times.Count Frequency130120110100908070100500Head CountCumulative PercentResults from 10,000 people doing a coin toss 200 times.Cumulative PercentWecannow equatea probability to theoccurrence of specific valuesorgroupsofvalues.Forexample,wecan seethat theoccurrence of a“Headcount” of lessthan74or

42、greater than124 outof200tosses is so rarethata singleoccurrence wasnotregisteredoutof10,000 tries.Ontheotherhand,wecan seethat thechanceofgettinga count near(or at)100ismuchhigher.With thedata thatwenow have, we canactuallypredict eachofthesevalues.COIN TOSSPROBABILITYDISTRIBUTION-6-5-4-3-2-10123456

43、NUMBER OF HEADSPROCESS CENTERED ON EXPECTED VALUE sSIGMA (s ) IS A MEASURE OF “SCATTER” FROM THE EXPECTED VALUE THAT CAN BE USED TO CALCULATE A PROBABILITY OF OCCURRENCESIGMA VALUE (Z)CUM % OF POPULATION586572798693100107114121128135142.003.1352.27515.8750.084.197.799.8699.99713012011010090807060050

44、04003002001000FrequencyIf we know where we are in the population we can equate that to a probability value. This is the purpose of the sigma value (normal data).% of population = probability of occurrenceCommonOccurrenceRare EventWHAT DOESITMEAN?What arethechancesthat this“justhappened”Ifthey aresma

45、ll,chancesarethatanexternalinfluenceisatworkthat canbeused to ourbenefit.ProbabilityandStatistics“theoddsofColoradoUniversity winningthe nationaltitleare3to1”“DrewBledsoespasscompletion percentagefor thelast6 games is .58%versus .78%for thefirst5 games”“The Senatorwillwinthe election with54% of thep

46、opularvote withamarginof+/- 3%”ProbabilityandStatisticsinfluenceour lives dailyStatistics is theuniversallanuageforscienceStatistics is theartofcollecting,classifying,presenting,interpretingand analyzingnumericaldata,aswell as makingconclusionsaboutthesystemfromwhichthedatawasobtained.Population Vs.

47、Sample(CertaintyVs.Uncertainty) A sample is just a subset of all possible valuespopulationsample Since the sample does not contain all the possible values, there is some uncertainty about the population. Hence any statistics, such as mean and standard deviation, are just estimates of the true popula

48、tion parameters.DescriptiveStatisticsDescriptiveStatisticsisthebranch of statisticswhichmost peoplearefamiliar.Itcharacterizes andsummarizesthemostprominentfeaturesofagivensetofdata(means,medians,standarddeviations,percentiles,graphs,tablesandcharts.DescriptiveStatistics describe theelementsofa popu

49、lationasawholeortodescribedata thatrepresentjust asample of elements fromthe entirepopulationInferential Statistics InferentialStatisticsInferentialStatisticsisthebranch of statisticsthatdealswithdrawing conclusions about apopulationbasedoninformationobtainedfrom asample drawn fromthatpopulation.Whi

50、ledescriptivestatistics hasbeen taughtforcenturies,inferentialstatistics is arelativelynewphenomenonhavingitsrootsinthe 20thcentury.We“infer” somethingabouta populationwhenonly informationfrom asample is known.ProbabilityisthelinkbetweenDescriptiveandInferentialStatisticsWHAT DOESITMEAN?-6-5-4-3-2-1

51、0123456NUMBEROFHEADSsSIGMAVALUE(Z)CUM%OFPOPULATION586572798693100107114121128135142.003.1352.27515.8750.084.197.799.8699.9971301201101009080706005004003002001000FrequencyAndthe first 50 trialsshowed“HeadCounts”greater than130?WHAT IF WE MADEACHANGETOTHE PROCESS?Chances arevery goodthattheprocessdist

52、ributionhas changed.Infact,thereisaprobabilitygreaterthan 99.999%thatithaschanged.USES OF PROBABILITY DISTRIBUTIONSCritical ValueCritical ValueCommon OccurrenceRare OccurrenceRare OccurrencePrimarilythesedistributionsareusedtotest forsignificantdifferencesindata sets.Tobeclassified as significant,th

53、eactual measured value mustexceed acriticalvalue.Thecriticalvalueistabularvaluedetermined by theprobabilitydistributionand therisk of error.Thisrisk of error is calledarisk andindicatesthe probability of thisvalueoccurring naturally.So,anarisk of .05(5%) means thatthiscriticalvaluewill be exceeded b

54、y arandom occurrencelessthan 5% of thetime.SOWHAT MAKES ADISTRIBUTION UNIQUE?CENTRAL TENDENCYWherea populationislocated.DISPERSIONHowwidea populationisspread.DISTRIBUTIONFUNCTIONThemathematical formulathatbest describesthedata(wewillcoverthis in detailinthenextmodule).COIN TOSSCENTRALTENDENCY1301201

55、101009080706005004003002001000Number of occurrencesWhat aresome of theways thatwecan easilyindicatethecentering characteristic of thepopulation?Threemeasureshave historicallybeen used; themean, themedianandthemode.WHAT IS THEMEAN?ORDERED DATA SET-5-3-1-10000013-6-5-4-3-2-101234564The mean has alread

56、y been used in several earlier modules and is the most common measure of central tendency for a population. The mean is simply the average value of the data.n=12xi=-2meanxxni=-=-21217.MeanWHAT IS THEMEDIAN?ORDERED DATA SET-5-3-1-10000013-6-5-4-3-2-101234564If we rank order (descending or ascending)

57、the data set for this distribution we could represent central tendency by the order of the data points.If we find the value half way (50%) through the data points, we have another way of representing central tendency. This is called the median value.Median ValueMedian50% of data pointsWHAT IS THEMOD

58、E?ORDERED DATA SET-5-3-1-10000013-6-5-4-3-2-101234564If we rank order (descending or ascending) the data set for this distribution we find several ways we can represent central tendency.We find that a single value occurs more often than any other. Since we know that there is a higher chance of this

59、occurrence in the middle of the distribution, we can use this feature as an indicator of central tendency. This is called the mode.ModeModeMEASURESOFCENTRAL TENDENCY,SUMMARYMEAN ( )(Otherwise known as the average)XXni=-=21217.XORDERED DATA SET-5-3-1-10000013-6-5-4-3-2-101234564ORDERED DATA SET-5-3-1

60、-10000013-6-5-4-3-2-101234564ORDERED DATA SET-5-3-1-10000013-6-5-4-3-2-101234564MEDIAN (50 percentile data point)Here the median value falls between two zero values and therefore is zero. If the values were say 2 and 3 instead, the median would be 2.5. MODE (Most common value in the data set)The mod