1、6-Sigma Training16-Sigma Training2Current position:Sr.Officer,LCM Eng.Dept MPhif.(EE),BEng(Hons)Eng.Past working experience:HKUST:To fabricate OLED and design the related driving cicruits.CityU:To fabricate Polymeric Opto WaveguidesPhilips:To develop LCM including B/W,Color STN,TFT and PLEDSolomon S
2、ystech:To develop the driver IC for B/W,Color STN,TFT and PDA products.6-Sigma Training3This course is to introduce the ways of Process Control System targeting to 6-sigma Quality in the manufacturing areas.The benchmark of Six Sigma will be shared during the course.In order to achieve it,series of
3、measurement system capabilities,performance metrics and process control methodology will be discussed.6-Sigma Training4 Participants will learn:-*What is six Sigma *Control Planning and Methodology *Measurement system capability *Process Capability Metrics *Continuous Process Improvement6-Sigma Trai
4、ning5*99.73%of the total area within spec.limits.Defect rate will be 2,700 PPM.*Only 93.32%of the area under the curve and defect rate is 66,000PPM.6-Sigma Training6*Defect rate-0.002 PPM*Defect rate-3.4 PPM6-Sigma Training7 Output VariablesManage the outputs.A Traditional ViewSales GrowthMarket Sha
5、reProfitability6-Sigma Training8 Output Variables Input VariablesManage the inputs;respond to the outputs.A Non-traditional ViewSales GrowthMarket ShareProfitabilityCustomer SatisfactionProduct QualityOn-Time DeliveryCOQCredit TermsRelationshipsServiceCustomer Training6-Sigma Training9Motorolas term
6、 to express a process capability that produces only 0.002 defects per each million units manufactured when the process is centered,or a maximum of only 3.4 defects for each million units when the process shifts up to 1.5 sigma to one side.The term“6 sigma”is becoming synonymous with“Quality Excellen
7、ce”.6-Sigma Training10 The Concept of Six SigmaDifferent numbers of Opportunities.Manufacturing ProcessesCustomers or SuppliersAdministrative Areas6=A structural approach to continuous improvement1-Identify the product or service you provide2-Identify your customers&their requirements3-Determine you
8、r needs&suppliers4-Define the process for doing work5-Eliminate defect sources/optimize the process6-Continuously improve the Sigma level(or“Six steps toward excellence”)A Bit of Statistics.A Bit of Statistics.-7.0-6.0-5.0-4.0-3.0-2.0-1.00.01.02.03.04.05.06.07.0upper spec limitLower spec limit+1.5 S
9、igma-1.5 Sigma3.4 ppm or Zero+6-Sigma Training11CENSORED1)(2nxxiNormal Distribution-Gaussian CurveSigma=Deviation(Square root of variance)-7-6-5-4-3-2-101234567Axis graduated in SigmaAxis graduated in Sigma68.27%95.45%99.73%99.9937%99.999943%99.9999998%result:317300 ppm outside(deviation)45500 ppm27
10、00 ppm63 ppm0.57 ppm0.002 ppm between+/-1between +/-2between +/-3between +/-4between +/-5between +/-6=6-Sigma Training12Six Sigma.And the Statistics -7.0-6.0-5.0-4.0-3.0-2.0-1.00.01.02.03.04.05.06.07.0+1.5 Sigma-1.5 Sigma3.4 ppmor ZeroUpper spec.limitLower spec.limit Reduce the variation:when =2 Cp
11、=2 andand Cpk =1.5 Cpk =1.5Cp =2Cp =26-Sigma Training13Six Sigma.And Capability 12345678Cp0.330.6611.331.6622.332.66Cpk-0.160.160.50.831.161.51.832.16Cp=Design specification width6 3 Cpk=Min.Mean-Spec LimitWith a maximum process shift of+/-1.5 6-Sigma Training14Different Numbers of Opportunities.Man
12、ufacturing ProcessCustomers or Suppliers Administrative Areas6-Customers or Suppliers:One opportunity per product delivered or per component purchased.-Manufacturing process:Opportunities at each process step)-Administrative areas:Number of opportunities for error for each activity performed The Opp
13、ortunities.6-Sigma Training15Some Examples of OpportunityCustomer Perspective:1 opportunity per product delivered ex.1 regulator=1 opportunity 1 controller=1 opportunity Production lines:Regulator =160 opportunities Controller=1200 opportunitiesForm(Payment,Purchase Order,.):Number of fields Expedit
14、e/Delivery:1 opportunity/packaging unit6-Sigma Training16A quality level of A quality level of 6 6 corresponds to less corresponds to less than 3.4 defects per million than 3.4 defects per million OpportunitiesOpportunities(i.e.,correct(i.e.,correct 99.99966%99.99966%of the time)of the time)Number o
15、f DefectsNumber of DefectsNumber of UnitsNumber of Units=D.P.UD.P.U D.P.U x 1000000D.P.U x 1000000Number of OpportunitiesNumber of Opportunities =D.P.M.Op.=D.P.M.Op.Sigma and Opportunities1 Opportunity=1 Opportunity=D.P.M.Op6-Sigma Training17A Universal Measurement Scale.3 34 45 56 67 76681066810621
16、062102332333.43.4 SigmaDPMOpDPMOpOn one condition:Calculate the defects and estimate the opportunitiesin the same way.6-Sigma Training18Measurement With SIGMA Is Simple!Estimate theOpportunitiesCount the Defects Follow the Indicator:Defects per million OpportunitiesConversion into Sigma canbe accomp
17、lished with the helpof a statistical table.6 6 =3.4 dpmo=3.4 dpmo6-Sigma Training19Converting Defect Levels to.Sigma!175 defects are identified whileproducing 5000 controllers The manufacture of one controllerallows for 1367 defect opportunities.D.P.Op=0.035/1367=0.0000256 An ExampleAn Example:D.P.U
18、=175/5000=0.035 D.P.M.Op=25.6 Sigma level :5.55Sigma level :5.556-Sigma Training20Six Sigma:An Ambitious Objective?Accurate to 99.99966%(less than 3.4 defects for each millionopportunities)could appear excessive.!99.9%is already VERY GOOD!But what could happen at a quality level of 99.9%(i.e.,1000 p
19、pm),in our everyday lives(about 4.6)?4000 wrong medical prescriptions each yearMore than 3000 newborns accidentally falling from the hands of nurses or doctors each yearTwo long or short landings at American airports each day400 letters per hour which never arrive at their destination6-Sigma Trainin
20、g21The Impact of Quality:“6 Sigma Suppliers”n13 wrong drug prescriptions per yearn10 newborn babies dropped by doctors/nurses per yearnTwo short or long landings per year in all the airports in the U.S.nOne lost article of mail per hour6-Sigma Training22Defect Reduction:“Peeling the Onion”Action Nam
21、eDateACTION PLAN PROCEDURECONTINUOUS IMPROVEMENT MODEL TREND CHART GOAL T I M E 100%BRAINSTORMING/PARETO ANALYSIS/ROOT CAUSE IDENTIFICATION RESULTS AND INSTITUTIONALIZE MEASURE ANALYZE PROBLEM SOLVE Processstep 1Processstep 3Processstep 2SUPPLIERSUPPLIERSUPPLIERCUSTOMERCUSTOMERCUSTOMERRECORDDEFECTSR
22、ECORDDEFECTSRECORDDEFECTSRECORDDEFECTSRECORDDEFECTS6-Sigma Training23The capability of the measurement system must be ensured before the system is used for process control or for measurements used in capability studies.Gage Repeatability&Reproducibility(R&R):is a statistical study to investigate the
23、 gages ability to repeat a measurement on the same part(repeatability)an dthe ability to give the same reading on the same part when used by different people(reproducibility).Under 10%-acceptable 10%-30%-May be acceptable depending on application.Effort should be made to improve the gaga capability.
24、Over 30%-Not suitable to use.6-Sigma Training246-Sigma Training25To measure the ability of a machine or process to produce a product within specification.Potential Process Capability Index(Cp):-Ratio of specification width over the process spreadSpec WidthCp=Process SpreadUSL-LSL =6-Sigma6-Sigma Tra
25、ining26Process Capability Index(CpK):-an off-target process should be“penalized”for shifting from where it should be.CpK is the index to measure this real capability when the off-target penalty is taken into consideration.Target-Process MeanPenalty/Correction factor,k=1/2 Specification Width Mean-Cl
26、oser spec limit Cpk=Cp(1-k)=3 sigma6-Sigma Training276-Sigma Training28*Maximum customer satisfaction is achieved only when the product is right on target.6-Sigma Training296-Sigma Training301.Process/Product Failure Mode and Effects Analysis(FMEA)An analytical technique aimed at problem prevention
27、through identification of potential problems.For each failure mode,the effect on the total system is studied and actions to minimize the probability of failures are planned.FMEA will serve:a)Identifies known or potential process failure modesb)Assesses the potential risk effects of the failures to t
28、he customerc)Identifies the potential product or manufacturing process causes and control actionsd)Identifies significant process variables to focus controls for detecting and reducing similar failures6-Sigma Training316-Sigma Training322.Process Control Plan A summary description of the procedures
29、for controlling all significant characteristics of the processes used to manufacture a family of similar products.It identifies the important processes used to produce a product,the significant characteristics associated with each process and how to control each of those significant characteristics.
30、Control Plan consists:a)Processb)Equipmentc)Significant characteristicsd)Measurement methode)Minimum sample size and frequencyf)Analysis methodg)Out of control action6-Sigma Training33QualityAfter etching qualityOKAbnormallotEngineer control6-Sigma Training343.Total Control Methodology A set of docu
31、ments providing all necessary instructions on how to control each significant variable in a process.A Total Control Methodology is used at the individual process level,while a Control Plan is used for a product line that contains several processes.Total Control Methodology consists:a)Positrol Planb)
32、Positrol Log check listc)Control Chartsd)Out-of-Control Action Plan(OCAP)e)Process/Machine Setup Check Listf)P/M Schedule and Checklistg)P/M procedures6-Sigma Training35CHART TYPEPURPOSEUSAGESUBGROUP SIZEREMARKSFraction Defective(p)Monitor the fraction of defects of a subgroup(units rejected/units i
33、nspected).Usually used at inspection points where one or several product characteristics are inspected result in an accept/reject decision for the unit.Variable subgroup size.Subgroup size must be large enough to have a high probability that at least one defect is present in the subgroup.Not effecti
34、ve when defect level is low and/or when the subgroup size is not large enough.Is not appropriate if a single defect will put the chart out of control.Alternatives,such as Positrol Log,should be considered in such situations.If actual subgrNumber of Defective(np)Monitor the number of Defects in a sub
35、group(number of units rejected out of sample inspected)Subgroup size is fixed,eliminating the need to recalculate the Control Limits for each subgroup.More convenient to plot than p chart.Subgroup size must be constant.Not effective when defect level is low;alternatives should be considered.Number o
36、f Defects per Subgroup(c)Monitor the number of defects in each subgroup.Used to monitor the number of defects in a unit.Subgroup size is one unit that could have several defects.Subgroup size must be constant.Ineffective when defect level is low;alternatives should be considered.Average Number of De
37、fects per Unit(u)Monitor the average number of defects in a unit.Used to monitor the average number of defects in a unit.Subgroup includes several unit and the size may vary.Control limits must be calculated individually for each subgroup if subgroup size varies.Alternatives should be considered whe
38、n defect level is low.6-Sigma Training36CHART TYPEPURPOSEUSAGESUBGROUP SIZEREMARKSAVERAGE(X-bar)Monitor the average of a process or product characteristic across time.Moderate to high volume continuous production when a sample of more than 3 units can be selected every few hours.n 3Insensitive to sh
39、ifts smaller than 2 signmas when n=4 to 6.Do not use run tests if data is autocorrelated.RANGE(R)Monitor the variability of a process or product characteristic across time.Used with a X-bar chart when sample size is 10 or smaller.3 n 10Increased sensitivity due to increased sample size.Recommended w
40、hen tighter control of a process is necessary and/or when sampling cost is not a factor.Do not use run tests.Z-BAR&WSimilar to X-bar&R chartsThese are standard X-bar&R charts,used to monitor the same characteristic of several similar,but not identical,products.3 n 10Non-Shewart chart.Sensitive to sm
41、all shifts in order of 0.5 to 1.5 sigma.Does no react to large shifts as quickly as Shehart X-bar chart.Able to forecast the process mean at the next period.More preferrable over CUSUM chart.CUMULATIVE-SUM(CUSUM)Similar to a EWMA chartSimilar to an EWMA chart,but difficult for manual charting.Recomm
42、ended only when computerized charting is available.n 1Non-Shewart chart.Sensitive to small shifts as an EWMA chart,but reacts slower than an EWMA chart to large shifts.6-Sigma Training37Aspects:Tester as a processTester as a gageProduct performanceMechanical equipmentOther operationTechniques:Cap.St
43、udies of forcing functionsMeasuring system analysisCap.Studies on product characteristicsCap.Studies on handlersCap.Studies such as tape and reelControl charts on output responsesSet-up verification using golden units as a hypothesis testControl charts on product characteristicsControl charts on han
44、dlers characteristicsControl charts onprocess characteristics Control charts on impact of handler on bent leadsDOE to optimize handler settingsDOE for process optimizationGage R&R on mech.Measurement systems such as optical comparators,lead scanners,etc.6-Sigma Training386-Sigma Training396-Sigma Tr
45、aining406-Sigma Training41=A structured approach to continuous improvement1-Identify the product or service you provide2-Identify the customer&their requirements3-Determine your needs&suppliers4-Define the process for doing work5-Eliminate defect sources/optimize the process6-Continuously improve the Sigma level(or”Six steps toward excellence)Six Sigma and Continuous ImprovementSupplierSupplierCustomerCustomerEnterNeeds ExitProduct/ServiceProcessActivity123456Feedback感谢观赏
