3 Quality Control Statistical Process Control or SQC Even if a process is in control, it may not be as capable as it could beFor example, the defect rate could be high, and this may be acceptable, but it may cost too muchWhich leads us to… Gryna
4 Quality Management Quality Planning Quality Control Quality Improvement Juran
5 Quality Improvement Types ContinuousIncremental improvement – small stepsIf done continually, major improvements occur over the long runBreakthroughMajor improvement at one timeCan allow leap-frog of competitors
6 Quality Improvement 1930’s and ‘40’s - Quality Control (QC) Post-WW II - Total Quality (TQ) and Toyota Production system (TPS)Total Quality Control (TQC)Total Quality Management (TQM) – 1980’s and ’90’s6 Sigma – 1985 to present
7 W. Edwards Deming Major source of poor quality is variation Quality improvement is the responsibility of managementAll employees should be trained in use of problem solving tools and statistical techniques.
8 W. Edwards DemingInvited to Japan after WWII to assist in reconstruction of industryEmphasized quality and variation reductionThe rest is history!Deming Prize introduced in Japan in 1951
9 6 SigmaUtilizes tools and techniques that span the range of all of quality managementSystematic improvement processProject drivenTrainingCertification – Green Belt, Black Belt, Master Black Belt
10 What is 6 Sigma? A statistical measure – standard deviation A measure of process capabilityA method of quality improvementGeneration I – variation reduction – 1980’sGeneration II – cost reduction – 1990’sGeneration III – customer focus – nowA method to design quality in – recentDFSS
11 Companies using 6 sigma Motorola – inventor, mid 1980’s GE Texas InstrumentsBank of AmericanCitibankBoeingHome Depotand many more
12 Six Sigma Defined“…a comprehensive and flexible system for achieving, sustaining, and maximizing business success.”“…driven by … understanding of customer needs, …use of facts, data, and statistical analysis, and … attention to managing, improving and reinventing business processes.”The Six Sigma Way – Pande, et al., p. xi
13 GE 1997 – 6000 projects, $320 million in savings. 1999 – $1.5 billion in savings1996 to 1999 – claimed a total of $4.5 billion in savings ~ 1.2% of revenue
14 Citibank Reduced credit processing time by 50% Reduced cycle times of processing statements from 28 days to 15 days
15 Bank of America Started using six sigma in 2001 Claimed ~$2 billion in benefits by 2003Many key customer processes near or at the 6 sigma level.Customer delight indicators up 25%Deposit processing improved by 47%
16 Standard Deviation Normal probability distribution Measure of dispersion of the dataCalculations
17 Measure of Process Capability Capability is the measure of how well the process performsUpper and lower specification limits – USL and LSLProduct is good within USL and LSLProduct is defective if outside the USL or LSLThe sigma level will be calculated using defects outside the USL and LSL
18 6 Sigma Levels and Defects Defects per million opportunities – DPMO2308,537366,80746210523363.4Assumes a 1.5 Z shift.
19 A Method of Quality Improvement Customer focusData and fact drivenProcess focus, management and improvementProactiveBoundary-less collaborationDrive for perfection, tolerate failure
20 Key Concepts of Six Sigma Critical to Quality: Attributes most important to customers.Defect: Failing to deliver what the customer wants.Process capability: What the process can deliver.Variation: What the customer sees and feels.Stable operations: Ensuring consistent, predictable processes to improve what the customer sees and feels.GE
21 5 Step Process - DMAIC Define the process and what customers require Measure the defects and the processAnalyze the data and discover causes of defectsImprove the process to remove causes of defectsControl the process to prevent loss of the improvements
22 Improving the ProcessTo reach desired sigma capability level, change the spec limits! Or…Reduce variationCalculate the USL and LSLCalculate the std dev, sigma, to reach the desired sigma capability level.
23 Calculating Sigma Levels Variables – any valueDiscrete (Attributes) – good/bad, count of defects
24 Variable Measures Normal Distribution USL and LSL Use Normal table to find % defectsFind the Defects Per Million Opportunities (DPMO)Use Table to Find Sigma Capability Level
25 Discrete TDU = total defects per unit TDU = total number of defects divided by total number of units sampledDPMO = defects per million opportunitiesOpportunities per unit = number of different possible defectsDPMO = TDU x 1,000,000 divided by opportunities per unitUse table to find Sigma capability level
26 Design for Six Sigma - DFSS Designing a new processOr a major redesign of an existing process5 step process – DMADVDefineMeasureAnalyzeDesignVerify