Surviving Six Sigma – What Statisticians Should Do To Not Be Left Behind Professor Tony Bendell Director, Centre of Quality Excellence, University of Leicester, UK and Managing Director, Services Ltd., Nottingham, UK Swiss Statistics Meeting ‘Quality and Statistics’ Aarau, Switzerland, 17-19 November 2004 Services Ltd., Quality & Reliability House, 82 Trent Boulevard, West Bridgford, Nottingham, NG2 5BL, UK Tel: +44 115 945 5285 Fax: +44 115 981 7137 E-mail: [email protected]www.servicesltd.co.uk
31
Embed
Surviving Six Sigma – What Statisticians Should Do To Not ... · Surviving Six Sigma – What Statisticians Should Do To Not Be Left Behind Professor Tony Bendell Director, Centre
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Centre of Quality Excellence, University of Leicester
Established January 2002Mission: To foster a rigorous, yet creative approach to the development of Quality Excellence in management, engineering, systems and services across the UK and world Industry, commerce and the public and voluntary sectors.In-company, Distance Learning and campus based MSc and postgraduateprogrammes in
- The Management of Quality Excellence- Customer Service Management
Broad research base – funding from EFQM, BQF, Rolls-Royce, DTI, Patent Office, LRQA, SGS Yarsley, EMDA etc.Recognition Centre for Six Sigma Black and Green BeltsSponsor and Associate organisations
Quality and Productivity Consultancy and Training organisation established 1983One of the three largest BQF licensed trainers for the EFQM Excellence Model and partner for Six SigmaIRCA licensed trainers for ISO 9001:2000Licensed by the British Accreditation Bureau etc.Specialisms:
Six Sigma - The Threat and the Opportunity for StatisticiansSix Sigma having major business impactTrains and supports its own peopleStatistically-based, but not just statisticsShould statisticians welcome or oppose Six Sigma? How should they get involved?
Extensively applied, growing application, high esteemManagement and Engineering press, conferences and growing academic literatureIndividual Black/Green belt first project annualised savings £10,000 - £1.5 million“The most important initiative GE has ever undertaken … it has taught GE employees about the importance of fulfilling expectations” Jack Welch
The Meaning of Six SigmaTechnical Term (Statistical)
12 Standard Deviations inside Specification“Best in Class”“three non-conformances out of one million opportunities”
Variation Reduction/Improvement approach that allows comparability between dissimilar processesMeasurement-based Corporate Strategy for improvement“Philosophy”
Given the specification, Six Sigma Quality for a product/process metric means that more of the distribution is contained within the Specification than at the Three Sigma level.
Six Sigma InfrastructureStrategic programmeProject-by-Project improvement in ‘fenced-off’ areasClear responsibilities and authority – Champions, Owners, Black and Green BeltsBefore and after performance measurement (typically cost)140+ statistical tools and conceptsDMAIC(T) methodology: Define, Measure, Analyse, Improve, Control, (Transfer)
Six Sigma Infrastructure Continued - Black and Green BeltsFull-time or Part-time (20%+) improvement engineers, managers, supervisors20/10+ days trainingExcel and Minitab (or alternative)Projects selected prior to trainingOn-site mentoringCertification
Important ToolsGE/Six Sigma Academy, 140+ statistical tools – to define, measure, analyse, improve and controlWhat are the key ones?
- Simple data description and summarisation tools- Regression and Anova- SPC and Measurement System Analysis- Experimental Design (including Taguchi methodology, Response Surfaces)- Quick technical problem solving
Non-statistical tools- FMEA/FMECA- QFD- Poka Yoke- Right brain and creativity tools
Not Just About Statistical ToolsProject-by-Project approach, clear metric and DMAICT crucialPrevious tool - use poor, disorganised, non-coherentMeasurement System AnalysisEmphasis on improvement, not Quality Assurance (e.g. attribute charting)Much so-called SPC is really SQC‘Control the X’s not the Y’s
- process variables not the product characteristicsExperimental Design key to successful projects
ISO 9001, 19944.20 Statistical TechniquesIdentification of NeedThe supplier shall identify the need for statistical techniques required for establishing, controlling and verifying process capability and product characteristicsProceduresThe supplier shall establish and maintain documented procedures to implement and control the application of the statistical techniques identified above
Some Other IssuesDanger of dilutionConcentration on Statistical techniques and left brain‘Old-fashioned’ statisticsLack of formal link to Policy DeploymentEmphasis on Cost Down, not Profit UpCulture/Buy-in/”hype”Integration with other Programmes
My ExperiencesVery effective focus programmeEven if infrastructure not right still effective projectsMay improve processes but not other infrastructureWhole programme should be planned and scopedVery good people development
Toolkit and Training IssuesLearning by roteTraining requirement – Black Belts without DOE!/Danger of dilution“Left-brain only” - 141 statistical tools/ old-fashioned statistics“Right-brain” creativity/innovation tools
Six Sigma and the StatisticianImpact of Japanese approaches in Six Sigma on traditional engineering and statistical roles, structures and work practices
Black Belts MAY be engineers or statisticiansNeed for mind-set change
- Fact-based Data-driven decision combined with Engineering knowledge and judgement
- Cost, Customer Focus and Business Decision Making focus- Balanced treatment of variation and risk
- enfranchisement of workforce in statistical methods- questions role of ‘statistical elite’- potential move from specialists to mentors, trainers and Project