摩托羅拉六西格瑪企業(yè)培訓(xùn) 2.5 Determine Process Performance

1、Additional Department InfoCopyright 2002 Motorola. All rights reserved.Six Sigma Black Belt Program 2.5 Determine Process Performance2.5 - 1Six SigmaBlack Belt Program2.5 - DetermineProcess PerformanceThese materials, including all attachments, are protected under the copyright laws of the United St

2、ates and other countries asan unpublished work. These materials contain information that is proprietary and confidential to Motorola University and are thesubject of a License and Nondisclosure Agreement. Under the terms of the License and Nondisclosure Agreement, thesematerials shall not be disclos

3、ed outsider the recipients company or duplicated, used or disclosed in whole or in part by therecipient for any purpose other than for the uses described in the License and Nondisclosure Agreement. Any other use ordisclosure of this information, in whole or in part, without the express written permi

4、ssion of Motorola University is prohibited.Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 22.5 Determine Process PerformanceObjectiveTo introduce Process Capability and the right method for calculatingSigma Performance. Calculate process sigma performance using t

5、heappropriate method.Key Topics Introduction to Calculating Sigma Performance Calculating Sigma Performance with Discrete Data Calculating Sigma Performance with Continuous Data2.1DetermineWhat toMeasure2.2ManageMeasurement2.4EvaluateMeasurementSystems2.3EvaluateVariation2.5DetermineProcessPerforman

6、ceAdditional Department InfoCopyright 2002 Motorola. All rights reserved.Six Sigma Black Belt Program 2.5 Determine Process Performance2.5 - 3Introduction to CalculatingSigma PerformanceCopyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 4Process Performance is Based

7、On .Voice ofCustomer Customer Requirements Voice of ProcessDefectivesY = CTQ / CTPCopyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 5Why do we do Process PerformanceCalculation? Document baseline performance. Provide direction to the project. Compare performance bef

8、ore and after.Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 6Steps for Calculating Sigma QualityLevelThe method to be used for calculating processperformance depends on the type data used: continuousor discrete.Steps in Calculating Sigma Performance1. Determine

9、the CTQs or CTPs.2. Determine type of data to be collected (continuous/discrete).3. Determine sampling plan.4. Gather appropriate data.5. Calculate Sigma Performance.Additional Department InfoCopyright 2002 Motorola. All rights reserved.Six Sigma Black Belt Program 2.5 Determine Process Performance2

10、.5 - 7Calculating Sigma PerformancewithDiscrete DataCopyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 8Calculating Sigma Performance -Discrete Data By examining the raw data, we can count thenumber of defects that do not meet customerrequirements and translate that

11、directly into adefect calculation referred to as Defects PerMillion Opportunities, or DPMO. Based on the DPMO, calculate the sigma qualitylevel.Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 9DPMO Defined DPMO = Defects Per Million Opportunities.=1M x DNO where:

12、D* = total number of defects counted in the sample: a defectdefined as failure to meet a Critical Customer Requirementor CCR. N = number of units of product or service. O = number of opportunities per unit of product or service fora customer defect to occur. M = million.* There must be at least 5 de

13、fects and 5 non-defects touse the DPMO formula.Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 10DPMO =1MUnits X OpportunitiesDetermine Sigma with DPMODefects_UnitsDefectsOpportunitiesMotorola. All rights reserved.2.5 Determine Process2.5 - 11Sigma Calculation Tab

14、leDEFECTS PERCAPABILITYSIGMAMILLION3.45.48.5132132487210815923333748368796813501866255534674661621081981072413903178642275028716359304456554799668078075796801115070135666158655184060211855241964274253308538344578382089420740YIELD99.9996699.9994699.9991599.998799.997999.996899.995299.992899.989299.98

15、4199.976799.966399.951799.931399.903299.86599.813499.744599.653399.533999.37999.180298.927698.609798.213697.72597.128496.40795.543594.520193.319391.924390.319988.49386.433484.134581.59478.814575.803672.574769.146265.542261.791157.926INDEX (Cpk)1.51.471.431.41.371.331.31.271.231.21.171.131.11.071.031

16、0.970.930.90.870.830.80.770.730.70.670.630.60.570.530.50.470.430.40.370.330.30.270.230.20.170.130.10.07LEVEL65.95.85.75.65.55.45.35.25.154.94.84.74.64.54.44.34.24.143.93.83.73.63.53.43.33.23.132.92.82.72.62.52.42.32.22.121.91.81.7460172Performance50000053.9828500.0301.6Copyright1.52002Copyright 2002

17、 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 12Discussion .Short Term DatavsLong Term DataCopyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 13Hotel Breakfast Delivery Case A hotel provides room service meals to its guests. Hotel policy is th

18、at the meal must be delivered at thetime scheduled by the customer. The hotel six sigma team has found from the Voice of theCustomer that a too early breakfast delivery willinconvenience the guest as well as a too late delivery. Guest research indicates that customers require that thebreakfast meal

19、be delivered within 10 minutes of thescheduled delivery time. The team designed a service that guarantees a breakfastmeal delivery within 10 minutes of the time scheduled byits guest.DPMO ExampleCopyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 14 Using the breakfas

20、t delivery example, lets calculate the DPMO andthe process sigma using this method from the data set on breakfastdelivery times:D = 205N = 725O = 1 (There is only one opportunity for a defect per breakfastdelivery. Either the delivery is within the customer limits of 10minutes or it is a defect.)DPM

21、O = Using the Sigma Calculation table, enter the DPMO column and lookup the process sigma directly. Sigma Quality Level Approximately 2.1= 282,586205 (10 )6725x1DPMO ExampleAdditional Department InfoCopyright 2002 Motorola. All rights reserved.Six Sigma Black Belt Program 2.5 Determine Process Perfo

22、rmance2.5 - 15Calculating Sigma PerformancewithContinuous DataCopyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 16Process Capability Study A process capability study is one of the major steps ofthe continuous improvement process. It is part of anoverall strategy of

23、Six Sigma and process improvementthat has three objectives: Obtain Stable processes. Reduce the Variability of key process outputs. Improve the Capability of key processes through the reductionof variation and the centering of the process on its targetvalue. The capability of a process is increased

24、relative torequired tolerances or process specifications byreducing the variation in the process and center inprocess variables on their respective targets.Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 17Process Capability StudyA process capability study general

25、ly consists of four steps:Step 1.Step 2.Step 3.Step 4.Verify that the process is stable.Determine if the data distribution is normal.Calculate the Capability Indices Cp and Cpk;determine Sigma Quality Level.Make recommendations for processimprovement.Copyright 2002 Motorola. All rights reserved.2.5

26、Determine Process Performance2.5 - 18Step 1: Monitoring ProcessStability - Trend Charts Process Stability: The distribution characteristics ofthe measurements (e.g. location, spread, shape)remain constant over time. Trend Charts: Time ordered plots of data demonstratethe stability of the distributio

27、n of measurements overtime. Control Charts: A special case of a Trend Chart thatincludes data based control limits. Control Charts arethe primary tools for monitoring the stability of aprocess. Control Limits used to objectively indicate when aprocess has become unstable (or out of control).Copyrigh

28、t 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 19Step 1: Monitoring ProcessStability - Trend ChartsExample 1:Trend plot with the datapoints connected. Do you see anytrends? Is this process stable?020040060080010001200140016001800200017.5317.5217.5117.5017.4917.4817.4717.

29、4617.4517.4417.4317.4217.4117.4017.3917.3817.37Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 20Step 1: Monitoring ProcessStability - Trend ChartsExample 2:Trend plot of 100measurements takenonce per shift overseveral weeks. Are time trendsindicated in themeasure

30、ments? Is this process stable?05101520253035404550556065707580859095100170165160155150145140135130125Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 21Step 2: Determine if the DataDistribution is Normal In process capability studies, the correctinterpretation of t

31、he capability indices (Step 3)requires that the underlying measurementshave approximately a normal distribution.Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 222600270028002900300031003200330034003500360002500642Normal Dist.LSL (2700)USL (3300)Nominal (3000)Step

32、 3: Assessing ProcessCapability - Capability IndicesA capable process is one where most of the populationmeasurements fall inside the lower and upperspecification limits. = 1708Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 23Capability Indices Improving process

33、capability is one of the threenecessary requirements of a sound SPC strategy forquality improvement (in addition to obtaining stableprocesses and reducing variability). Capability is defined as the ability of a process toproduce outputs that meet engineering and/orcustomer specifications. A capable

34、process is one where the distributions of theprocess output measurements are centered on thetarget, and a very high percentage of themeasurements fall within the specification limits. Capability indices are introduced as a means ofmeasuring the capability of a process.Copyright 2002 Motorola. All ri

35、ghts reserved.2.5 Determine Process Performance2.5 - 24Uses of Capability IndicesCapability indices can be used to provide: A method of tracking the relative improvement of anindividual process over time. A method for estimating the percentage of defects or non-conforming product. A means of compari

36、ng the capability of several processes,each with different units of measurement and differentspecifications. A means for identifying the processes most in need ofimprovement. One set of acceptance criteria for transferring a processfrom a development area to a manufacturing line. One set of qualific

37、ation criteria for assessing suppliers.Copyright 2002 Motorola. All rights reserved.2.5 Determine Process PerformanceDefinition of CpDefinition of CpCp =Allowable Process VariabilityActual Process VariabilityCp =Cp =(population)(sample)USLUSL LSL6 USL LSL6 sAllowableLSLActual2.5 - 25Copyright 2002 M

38、otorola. All rights reserved.2.5 Determine Process Performance2.5 - 26Interpretation of CpLSLUSL 2.0Motorola 6CapabilityCp 1.5InterpretationPoor CapabilityMarginalCapablityGood Capability2.5 Determine Process PerformanceLSLCpk = 2.0T2.5 - 27Cpk = 1.0USLCopyright 2002 Motorola. All rights reserved.Cp

39、k = 1.0Three Processes with Cp = 2.0USL xbar xbar LSL Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 28Definition of CpkCapability index - Cpk Cp does not take into account the closeness of themean to the “target”. Cp by itself is insufficient to describe the cap

40、ability ofa process to conform to specifications. An index that does take into account where the meanof the sample is relative to the specification limits isCpk.Cpk = min,3s 3s2.5 Determine Process Performance2.5 - 29 One Sided Specification - Lower Limit (LSL)Definition of Cpk: One Sided Specificat

41、ionDefinitions of Cpk - One Sided Specifications One Sided Specification - Upper Limit (USL)Cpk = C pu = USL - 3 =Cpk = C pl - LSL3 USLCopyright 2002 Motorola. All rights reserved.LSLCopyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 30Assessing ProcessCapability - C

42、apability Indices If the distribution of measurements iscentered on the target(i.e., xbar = target), then Cpk = Cp.Otherwise, Cpk 2.0 and aCpk 1.5.Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 31What Is the Cpk for the Data in This Figure?13514014515015516016525

43、20151050LSL (145)USL (165)Nominal (155)StDev = 3.4115Mean = 150.26C pkC pk=m in ( 1 6 5 - 1 5 0 . 2 6 ) /3 ( 3 .4 1 1 ) , ( 1 5 0 .2 6 - 1 4 5 ) / 3 ( 3 .4 1 1 ) m in 1 .4 4 , 0 . 5 1 3 0 .5 1 3Copyright 2002 Motorola. All rights reserved.2.5 Determine Process PerformanceT2.5 - 32LSLUSLCombined Use

44、of Cp and CpkFour Processes with Cpk = 1.5D:Cp = 6.0C:Cp = 3.0B:Cp = 2.0A:Cp = 1.5Distribution| T |kCpCpkCpk/CpA0.001.330.001.501.501.00B1.501.000.252.001.500.75C3.000.670.503.001.500.50D4.500.330.756.001.500.25Copyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 33Com

45、parison of Cp and CpkCopyright 2002 Motorola. All rights reserved.2.5 Determine Process Performance2.5 - 34Common Mistakes with CapabilityIndices Calculating indices on an unstable process. Calculating standard indices when the distribution isnot normal. Specifications must be meaningful. Using too small a s