Six Sigma 6s
Dilbert
Introduction 6 : new world s is the symbol for the standard deviation. 6s is equivalent with 3.4 defects per million opportunities. Hooving 30 m2 on 6s -level means only 1 cm2 missed. 1/3.4 million part of the day equals 0.29 second 1/3.4 million part of the equator of the earth equals about 140 meter.
Statistical background e r a n c e L S L U S L + / - 3 s 1 3 5 1 3 5 p p m p p m . 1 . 1 p p m p p m N o m i n a l = m + / - 6 s
Statistical background e r a n c e L S L U S L 1 . 5 s 3 . 4 6 6 8 3 p p m p p m p p m 3 . 4 p p m m + / - 6 s
What is Six Sigma? Simple Complex Eliminate defects Eliminate the opportunity to have defects Complex Vision Metric (Standard measuring method) Benchmark Philosofy Method Tool for: Customer satisfaction ‘Breakthrough’ improvements Continuous improvement Employee involvement Agressive goals
Six Sigma is everywhere Service Design Management Purchase Six Sigma Methods Administration Production IT Quality Depart. HRM M & S
Uniform standard 2 3 4 5 6 308537 66807 6210 233 3.4 PPM 69.1% 93.3% ‘Sigma’-metric 2 3 4 5 6 308537 66807 6210 233 3.4 PPM 69.1% 93.3% 99.38% 99.977% 99.9997% Yield Process performance Defects per million Long term yield Current standard World class
Financially interesting Benefits of 6s approach w.r.t. financials
Why improvement? Process step Inspection Client The hidden factory Input Process step Inspection good products Client Redo Waste The hidden factory
Companies implementing Six Sigma Motorola (1987) Texas Intruments en DEC (1988) IBM (1989) AlliedSignal en Kodak (1994) General Electric (1995) Whirlpool, Bombardier en Polaroid (1996) Sony, Citibank (1997) Paccar-DAF Trucks, Black & Decker (1998) Perlos, American Express (1999) Ford (2000)
A scientific and practical method to achieve improvements in a company Six Sigma A scientific and practical method to achieve improvements in a company Scientific: Structured approach. Assuming quantitative data. Practical: Emphasis on financial result. Start with the voice of the customer. “Show me the data” ”Show me the money”
Method Define Measure Analyse Improve Control DMAIC Select a project Make concrete the project Analyse Characterise the current situation Improve Optimise the process Control Assure the improvements DMAIC
Method Y = f(X1, X2, …, Xn) Focus on the process: Output Inputs Caffeine% = f(DCM, time to extract, number of extracts, temperature, …)
New way? Y = f(X1, X2, …, Xn) Approach to improve Praktical problem Statistical problem Y = f(X1, X2, …, Xn) Praktical solution Statistical solution
Define Projects may be selected according to: A complete list of requirements of customers. A complete list of costs of poor quality. A complete list of existing problems or targets.
Priorities 1 2 3 1.Collect data 2.Arrange the information 3.Give priority Financial benefits Expected throughput time of the project Severity of the problem 1. Requirements, 2. Costs, 3. Problems.
Characteristics involved Six Sigma Characteristics involved Y = f(X1, X2, …, Xn) Critical to Quality (External) Critical to Quality (Internal) Influence factors: Process parameters Disturbances Noise variables
Example Coffee beans Roast Cool Grind Moisture content Pack Sealed
Measure Measure Concrete the problem: Establish how to measure in the process; Investigate the demands (USL, LSL); Evaluate the measurement method. Voorbeelden: Treinen te laat: wanneer is een trein te laat? Wat wordt bedoeld met: 83% van de treinen is te laat? Vocht%: de EK is: vocht% in het pak (= vanuit perspectief van de klant). IK: voor het malen. Eisen: onze kritische vragen hebben er uiteindelijk toe geleid dat men de USL verhoogd heeft naar 4.8% (werd pas duidelijk toen wij in het laatste blok waren).
Measure USL 4.0 4.5 Moisture%
Analyse Analyse Study the current performance of the process: Determine the kind of problem; Identify the influence factors.
Analyse To run experiments one SPC needs a process which is in statistical control. SPC DoE
Uncontrolled processes Control chart, Cp, Cpk Process Capability study Uncontrolled processes Controlled processes Identify X’s FMEA, Control chart Multi-vari, Trendgraph Identify X’s Select X’s FMEA, Data Testing- theory Select X’s Improve Improvements Experiment Assure, Control loop Control
Analyse Waste! Wat moet er met dit proces gebeuren?
Analyse Uitschieters koppelen aan: wat gebeurde er op dat moment in het proces? -> stagnatie van het aftransport. Deze fase van het project is een soort detective spelen: meten -> kritisch bekijken -> mogelijke verklaringen zoeken. Toetsen van de mogelijke verklaringen is nog niet aan de orde. Belang van meten: als je niet meet, ga je je ook geen nieuwe vragen stellen, maar blijf je in het oude systeem rondhangen.
Analyse USL USL
Undesirable sources of variation. Three types of influence factors Caffeine content of Extracted coffee Time to extract Caffeine% raw coffee Stuffed-up filter Processparameter Noise variable Disturbance May be used to control. Undesirable sources of variation. Outliers, incidents
Improve Improve Run a statistical experiment: Determine which influence factors are really important; Model the relationship with the CTQ.
Moisture% = –16.57 + 0.023B + 0.025G – 0.0003BG Improve Moisture% = –16.57 + 0.023B + 0.025G – 0.0003BG
Control Control Implement improvements: Use the relationship to define an optimal process; Design a system to assure the improvements.
Control Before After σ = 0.04 USL σ = 0.19
Organisation
Persons Champion Belongs to management Selects projects Gives priority Solves bottlenecks Is willing to change Has social skills 1 day training Is responsible for the BB project w.r.t. the organisation
Improvement potential: € 50 000 Persons Black Belt Belongs to the middle management Is well-educated Project is related to daily activities May prioritise his work Enough motivation Willing to change Has good social skills Improvement potential: € 50 000
Throughput time project BB training Define Throughput time project Measure 4 months (full time) Analyse Training (1 week) Work on project (3 weeks) Review Improve Control