Document

Report
Six Sigma
Operations Management
Dr. Ron Lembke
6 sigma


DPMO: Defects Per Million Opportunities
DMAIC: Define, Measure, Analyze, Improve,
and Control


(Alternate meaning: Dumb Managers Always
Ignore Customers)
DCDA: Plan, Do, Check, Act
6  (6 sigma)



The goal is to ensure that no unacceptable parts are
ever passed on to a customer.
A defect is anything that does not fall within the
customer’s tolerance limits
Through continuous process improvement,
 Lower the process variability so low that the
upper and lower specifications are 6 standard
deviations above and below the mean
6  (6 sigma)
3
3 sigma:
Probability outside range = (1 – 0.99865) * 2 = 0.0027
Defect rate = 2,699 defects per million opportunities
6 sigma:
6
Probability part outside range = 0.00000000198024
Defect rate = 0.00197 dpm
1.97 defects per BILLION
Defect Rates - 1






3 sigma: 1/.0027 = 1 every 370 parts
6 sigma: 1/ 0.00000000198024
= 1 every 504.9 million parts
If we make a million parts per year, we have:
3σ: 2,699
defectives
6σ:
0.0019732 defectives
Defects - 2




With a 1.5σ shift, defect rates become:
3σ 66,807 dpmo
6σ 3.4 dpmo
The commonly accepted definition of 6σ
quality is having a defect rate <= 3.4 dpmo
1.5
3
4.5
6
Black Belts




Green Belts: some 6 sigma
training, take part in teams,
small solo work
Black Belts: Coach or lead 6
sigma improvement teams
Master Black Belts: have indepth statistical training, serve
as Black Belts for more teams
Champions: Executives who
will back up the proposals the
black belts come up with
Pareto Chart - ranked histogram


Invented by Joseph Juran
Beer defects
2.5%
2.0%
1.5%
1.0%
0.5%
0.0%
Sediment
Hoppy
Flat
Skunky
Misc
Defects
Wilfredo Pareto 1848-1923




Italian Economist
“80/20” rule: 80% of the wealth is
controlled by 20% of the people
Cours d'économie politique (1896-7)
80/20 rule believed to apply much
more widely
1906- “Pareto Optimality” – not
possible to make anyone better off
(in his own estimation) without
making someone else worse off
Cause & Effect Diagram Example
Method
Drill
Manpower
Over
Time
Slow
Tired
Wood
Not dried
Material
Too Many
Defects
Not
maintained
Steel
Lathe
Machinery
Control Chart Example
70
X
60
UCL
50
40
30
20
LCL
10
0
1
2
3
4
5
6
Time
7
8
9
10
11
12
Dilbert’s View
Fortune Story

58 large companies have announced Six
Sigma efforts
91% trailed S&P 500 since then, according to
Qualpro, (which has its own competing
system)

July 11, 2006

Qualpro’s “Six Problems with Six Sigma”

Qualpro’s “Six Problems with Six Sigma”






Six sigma novices get “low hanging fruit” “Without years
of experience under the guidance of an expert, they will not
develop the needed competence”
Green belts get advice from people who don’t have
experience implementing it
Loosely organized methodology doesn’t guarantee results
(and they do?)
Six Sigma uses simple math – not “Multivariable Testing”
(MVT)
Six Sigma training for all is expensive, time-consuming
Pressure to “do something” – low value projects
Qualpro Errors


Correlation is not causality
“Six Sigma is merely an overpriced set of low-level statistics
courses”





Stock price is the only important measurement?
How were 58 companies selected?
What was date of starting Six Sigma?
Comparing 12 MVT vs 12 Six Sigma:


Charles Holland, CEO, ChiefExecutive.net 1/29/07
84% vs 95% since 9/17/01
DuPont: MVT saved us $23 million

Six Sigma saved us $3 billion 3/07
Six Sigma


Narrow focus on improving existing
processes
Best and Brightest not focused on developing
new products

Fortune July 11, 2006

Can be overly bureaucratic
Summary




What is Quality?
Pioneers of Quality
ISO 9000, and Malcolm Baldrige National
Quality Award
Six Sigma



Much in common with TQM
DMAIC, Green and Black Belts
Six Sigma isn’t perfect, but very useful

similar documents