《six sigma training materials 六西格玛管理培训材料.ppt》由会员分享，可在线阅读，更多相关《six sigma training materials 六西格玛管理培训材料.ppt（101页珍藏版）》请在三一文库上搜索。
1、Six-Sigma,Six - Sigma Training Material,What is Six-Sigma,Key Terms,Six Sigma A term coined by Motorola to express process capability in parts per million.A six sigma process generates a defect probability of 3.4 parts per million (PPM) ChampionAn upper level business leader who facilitates the lead
2、ership,implementation,and deployment of six sigma philosophies. Black BeltA process improvement project team leader who is trained and certified in six-Sigma methodology and tools and who is responsible for project execution. Mater Black BeltA person who is an “expert ” on Six Sigma techniques and o
3、n project implementation,master Black Belt play a key role in training and coaching of Back and Green Belts. Green BeltSix Sigma role similar in function to Black Belt but length of training and project scope are reduced to two weeks of training.,Key Terms,Yellow BeltHourly personnel trained in the
4、fundamentals of six-sigma who assist and support in project execution ,usual work with black and Green Belt. Process MapA step-by-step pictorial sequence of a process showing process inputs,process outputs,cycle time ,rework operations,and inspection points. Key process Inputs Variable (KPIV) The vi
5、tal few input variables,call “x”s Key process outputs Variable (KPOV) The output variables,call “x”s DFMADesign for manufacture and assemble, A methodology to reduce product complexity and design around more capable components / processes Cost of poor qualityCost associated with Providing poor quali
6、ty products or services. Can be divide into four cost categories :Appraisal ,Scrap, Rework,and field Complaint,What is Six-Sigma,What is Six-Sigma Vision Philosophy Aggressive Goal Metric (Standard of measurement ) Benchmark Method Tools for : Customer focus Breakthrough Improvement Continues improv
7、ement People Involvement Six-Sigma is a problem solving process used- to produce: Reduced variation in our processes / Products improved RTY , DPU, Determine how each selected input varable can go wrong and place that in the Failure Mode column of the FMEA.,C a faulty temperature sensor. Identificat
8、ion of Causes should start with Failure Modes associated with the highest severity ratings. Examples Temperature Too High:Thermocouple out of calibration Surface Contamination: excess flux from hand soldering Incorrect PO number: Typographical error Pits: High particle count in clean room paint to t
9、hin:High solvent content,Failure Mode and Effects Analysis,3. Identify Potential Causes of each Failtre Mode As in most cases,we have several causes for one failure mode effect combination,Definition of Terms Current Controls Current Controls Systematized methodsdevices in place to prevent or detect
10、 failure modes or Causes (before causing effects).,Failure Mode and Effects Analysis,Definition of Terms Current Controls ( Continue ) Prevention consists of fool proofing, automated control and set-up verifications Controls consists of audits, checklists, Inspection,laboratory testing, training, SO
11、Ps,Preventive maintenance, etc. Which is more important to process improvement, Prevention or detection? Examples Thermocouple ort of calibration: PM of thermocouple Excess flux from hand soldering: Operator training Typographical error: Spell grammar proof software High particle count in clean room
12、: None High solvent content: None,Failure Mode and Effects Analysis,4.List the current Controls for each Cause For each Failure Mode/Cause we list how we are either preventing the Cause or detecting the Failure Mode ,we will list the procedure number where we have a SOP,This is how the FMEA identifi
13、es initial holes in the Current Control Plan-process teams can start working on these holes right away,Failure Mode and Effects Analysis,Risk Priority Number ( RPN ) The output of an FMEA is the “Risk Priority Number” The RPN is a calculated number based on information you provide regarding the pote
14、ntial faire modes, The effects, and The current ability of the process to detect the failures before reaching the customer It is calculated as the product of three quantitative ratings, each one related to the effects,causes, and controls: Risk Priority Number is not sacred Scaling for Severity, Occ
15、urrence and Detection can be locally developed. Other categories can be added.For example, one Black belt added an Impact score to the RPN calculation to estimate the overall impact of the Failure Mode on the process.,Failure Mode and Effects Analysis,FMEA Model,Cause,Detection,Failure Mode (Defect)
16、,Detection,Effect,Detection,Prevention,Control,Material or Process input,Process Step,External customer or downstream process step,Which is a best case? Which is a worst case?,Failure Mode and Effects Analysis,Definition of RPN Terms RPN = Severity X Occurrence X Detection Severity (of Effect)- impo
17、rtance of effect on customer requirements could also be concerned with safety and other risks if failure occurs (1= Not Severe, 10=Very Likely) Occurrence (of Cause) frequency with which a given Cause occurs and creates Failure Mode. Can sometimes refer to the frequency of a Failure Mode (1=Not Like
18、ly, 10=Very Likely) Detection (capability of Controls) ability of current control scheme to detect or Prevent: the causes before creating failure mode the failure modes before causing effect 1=Likely to Detect, 10= Not Likely at all to Detect,Failure Mode and Effects Analysis,FMEA Scoring There are
19、a wide variety of scoring “anchors”, both quantitative or qualitative. Two types of scales are 1-5 or 1-10. The 1-5 scales make it easier for the teams to decide on scores The 1-10 scale allows for better precision in estimates and a with variation in scores. The 1-10 scale is generally considered t
20、o be the best option.,Failure Mode and Effects Analysis,Example Rating Scale,Failure Mode and Effects Analysis,5. Assign Severity,Occurrence and Detection ratings to each Cause 6. Calculate RPNS 7. Determine Recommended Actions to reduce High RPNs What to do about the high RPNs * Review results and
21、insights * Determine potential next steps: Data collection / Experiments / Process improvement / Process control implementations Actions are recommended for only the high RPNs ( The key is FOCUS! ) 8. Take appropriate Actions and Document,Failure Mode and Effects Analysis,9. Recalculate RPNs, The FM
22、EA should be re-evaluated by the group as new recommended actions Are identified,completed and recorded.,Failure Mode and Effects Analysis,Methodology Two major approaches: Starting with QFD / Cause used as an estimate of short-term variation. Estimated by the pooled ( average ) standard deviation o
23、f the distribution of repeated measurements Reproducibility The different in the average of the measurements made by different persons using the same or different instrument when measuring the identical characteristic. Good reproducibility : the shift among different operators are little. Poor repro
24、ducibility : the shift among different operators are more.,MSA - GR&R,Precision to tolerance Ratio P/ T=5.15*MS / Tolerance Addresses what percent of the tolerance is taken up by measurement error. Best case : 10%Acceptable : 30% Includes both repeatability and reproducibility Operator x Unit x Tria
25、l experiment The P/T ratio(% Tolerance In Minitab) is the most common estimate of Measurement system precision. This estimate may be appropriate for evaluating how well the measurement system can perform with respect to specifications. Specifications,however,may be too tight or too loose. Generally
26、,the P/T ratio is a good estimate when the measurement system is only used to classify production samples. Even then, if process capability ( Cpk ) is not adequate, the P/T ratio may give you a false sense of security.,MSA - GR&R,%R&R= MS / Total x 100% Addresses what percent of the total variation
27、is taken up by measurement error. Includes both repeatability and reproducibility Operator x Unit x Trial experiment As a target, look for %R&R 30% The %R&R is the best measure for the process improvement leader This estimates how well the measurement system performs with respect to the overall proc
28、ess variation %R&R is the best estimate when performing process improvement studies.,If Cp0 =1.5 , then %R&R 30%,note,MSA - GR&R,Still Other statistical Indexes The Signal-Noise Ratio ( S/N Ratio ) relates the product variation to the measurement system variation . The S/N,S/N Ratio =,The Discrimina
29、tion Index provides the number of divisions that the measurement system can accurately measure across the part ( sample ) variation ,if this index is less than 2 ,then it is inadequate to provide data for a study. If the index is 2 ,then it is equivalent to a go / no-go gage. We would like to see th
30、e value 10.,Discrim =,MSA - GR&R,Other Issues Number of operators If process use multiple operators ,choose 2 - 4 at random If process uses only one operator ,or no operator , perform study without operator effects ( reproducibility effects ignored ) Number of samples Select enough samples so that (
31、 number of samples ) X ( number of operators ) 15 If not practical or possible , choose number of trials so that : if S * O 4 , trials = 6 if S * O 5 , trials = 5 if S * O 8 , trials = 4 if S * O 15 , trials = 3,MSA - GR&R,Study Sample Selection Samples should be pulled from the process that span th
32、e normal variation of the process Example : if you produce a material with a thickness specification of 0.010 +/-0.002, get samples that go from 0.008 - 0.012 thick. Be careful ! If you produce different thickness of material with the same process, Sub-group them and perform the R&R study .,MSA - GR
33、&R,Typical Gage R&R study Form,MSA - GR&R,Doing the problem Using the Minitab Macro Create A Minitab Data Set : One column for operator one column for Trials One column for sample for part One column for the output Variable Run the Minitab macros In Stat Quality Tools Gage R&R Study OR Gage Run Char
34、t ( Number of Distinct Categories must be at least 4 for Process Improvement use ),MSA - GR&R,SPC,How Do We Manage Data - Today SPC S = Statistical techniques used to examine process variation P = Process ,ANY Process C= Controlling the process through active management Where Did SPC Come From ? 192
35、0s - Western Electric / Dr . Walter shewhart Used to identify controlled & uncontrolled variation also known as common & special causes Tries to find the process signals in all of the noise Uses control chart as main tool,Cause Common Cause ( Noise ) Is present in every process Is produced by the pr
36、ocess itself Can be removed and / or lessened but requires a fundamental change in the process Special Cause ( signals ) Exists in most operations / process Caused by unique disturbances or a series of them Can be removed / lessened by basic process control and monitoring,SPC,Two general kinds of Da
37、ta Attribute - The data is discrete ( COUNTED ). Results From using go / no-go gages,or from the inspection of visual defects , visual problems, missing parts, or from pass / fail or yes / no decisions Variables - The data is continuous ( measured). Results from the actual measuring of a characteris
38、tic such as diameter of a hose,electrical resistance , ect.,SPC,Choosing the correct control chart,What type of data ?,Continuous,Data collected in groups or individuals ?,X-Bar R X-Bar S,Groups (Averages ),Individuals moving range,Discrete,Counting specific defects or defective items ?,Is the proba
39、bility of A defect Low Poisson distribution,Specific types of “Defects”,If you know how many are bad ,do you know how many are good? Binomial distribution,Defective Items,Area of Opportunity constant in each sample size,yes,Individuals moving range,No,No,Constant sample size ?,yes,U-Chart,C-Chart,No
40、,yes,P-Chart,NP-Chart,No,yes,SPC,Detecting Lack of control “Rules of the road” Start with rule #1 and pattern Detection rules If high sensitivity is need , go with #2,3,&4 Rule#1 : One point outside the UCL or LCL ( 3-sigma limit ) Rule#2 :Two of three consecutive outside the 2-sigma limit Rule#3 :
41、Four of five consecutive outside the one-sigma limit Rule#4 : Eight consecutive on one side of the center line Pattern Rule : A Pattern repeats itself Run the Minitab macros In Stat Control Charts X-Bar - (Open Tests choose Rules ),SPC,Capability,Steps to a Capability Study 1. Set up the process to
42、your “ best guess ” best setup and record the values of you key process input variables. 2. Identify a reasonable way to create rational subgroups. 3. Run the product over a short period of time to minimize the impact of special cause variation as possible .Approximately 30 time points are the targe
43、t for data collection. 4. Have your team carefully observe the process and take plenty of notes . 5. Measure and record values for the key process output variable,Steps to a Capability Study 6. Run capability Six-pack and review: Normal Plot SPC Charts ( check for stability,Accuracy) Histogram 7. Ru
44、n the Capability Macro for both the pooled and the overall standard deviations. Complete the worksheet. 8. Diagnose for Mean Shift and Variance inflation 9. Estimate Long - Term Capability 10 . Develop action plan based on diagnostics .,Capability,Capability summary Sheet Many time you will be inter
45、ested in the capability of more than one key output or key input variable.Use the capability summary sheet to track the status on the variables of interest .,Capability,Types of Capability Indexes Instantaneous Capability : Process Capability over an extremely short period of time . This should repr
46、esent the very best perform a process is capable of over a short time . This should be a close estimate of process entitlement . can be estimated using the “best run” of a short-term or long -term study. Short-Term Capability : capability study based on 3050 data points. Usually equal to or greater
47、than long -term capability long-Term Capability : Capability study bases on a large number of data points Best estimate of true process capability Diagnostics can be made from this data,Capability,Goal Setting This is a generic sequence of process improvement : Near Term Goal : Move the Ppk to Pp (
48、Center the process ) Mid Term Goal : Move the Pp to Cpk ( reduce Variation ) Long-Term Goal : Move Cpk to Cp ( Random Variation ) Answers to Capability Questions Describe dynamics and actions for the following scenarios : Cp 1.00 Inherent random variability is small enough for the process to consist
49、ently meet customer spec Cp 1.00 Inherent random variability to large . Process change required,Capability,Answers to Capability Questions Describe dynamics and actions for the following scenarios : Ppk Cp There are special cause shifts and drifts that drive variability over and above random variati
50、on.Improved control is the focus. Ppk is close to Cp Process in good state of statistical control Cp 1.00 and Ppk Cp Boy are you in trouble now,Eliminate special cause and investigate potential impact of APC (Automated process control ),Capability,Answers to Capability Questions ( Continued ) Cp 1.0