View
265
Download
2
Category
Preview:
Citation preview
RED BEAD EXPERIMENT Maarten Hoppen Donderdag 5 november 2015
W EdwardsDeming
David P.Joyce
Many thanks to!
Joakim SundénTwitter: @joakimsunden
joakim.sunden@spotify.comwww.joakimsunden.com
Joakim
Sundén
White Bead Company
The White Bead Co. IIP
ISO9000
WE LISTEN TO OUR PEOPLE
Only at their annual appraisal
CMMI 5
WE WANT YOU!
We Are Recruiting!3 willing workers. Must be willing to put in best efforts. Continuation of job is based on performance. No experience in making beads necessary. Educational requirements-nil.
2 inspectors. Must be able to distinguish red from white. Must be able to count to 20. No experience necessary.
1 chief inspector. Same qualifications as inspector. Must be able to speak in a loud voice.
1 recorder. Precise, follow orders, writing capability, human judgement
ABOVE AVERAGE WORKERS ONLY NEED APPLY
FINISHED FILES ARE THE RESULT OF YEARS
OF SCIENTIFIC STUDY COMBINED WITH
THE EXPERIENCE OF MANY YEARS.
SHOUT OUT YOUR COUNT
3 - 2 - 1 - SHOUT
FINISHED FILES ARE THE RESULT OF YEARS
OF SCIENTIFIC STUDY COMBINED WITH
THE EXPERIENCE OF MANY YEARS.
Lessons learned • Did incentives work?
• Did the motivational posters help?
• Did have a CMMI standard process work?
• What could managers do to improve things for the workers?
• What are the red beads in your company?
Lessons learned • It’s the system not the workers • It’s management thinking that designed the system • Arbitrary numerical targets were completely ineffective • Rewarding or punishing the workers had no effect • Rigid and precise procedures are not sufficient to produce
quality • Keeping the ‘best’ workers did not work
Management tampering creates more problems than it solves • Posters and slogans are at best useless and can be insulting and
create resentment • The biggest source of variation was in the system
Theory of Variation
The First Principle of the Theory of Variation
We Should Expect Things toVary,
They Always Do
My performancePe
rfor
man
ce
Time
✓
✗
My performance✓
My performancePe
rfor
man
ce
Time
✓
✗
My performancePe
rfor
man
ce
Time
✓
✗
My performancePe
rfor
man
ce
Time
✓
✗
My performancePe
rfor
man
ce
Time
✓
✗
My performancePe
rfor
man
ce
Time
✓
✗
My performancePe
rfor
man
ce
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
My performance
Perf
orm
ance
Time
✓
✗
Statistical Process Control Charts
Perf
orm
ance
Time
✓
✗
Upper Control Limit
Lower Control Limit
Average
Statistical Process Control Charts
The Second Principle of the Theory of Variation
Understanding Variation Will Tell us
What to Expect
Signal or Noise?
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Velocity
ControlLimits
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Signal or Noise?
Velocity
Reward!
No more Mr Nice Guy!
Managerrepents...
Tough managementworks!
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Signal or Noise?
Velocity
Reward!
No more Mr Nice Guy!
Managerrepents...
Tough managementworks!
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Signal or Noise?
Velocity
Reward!
No more Mr Nice Guy!
Managerrepents...
Tough managementworks!
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Signal or Noise?
Velocity
Reward!
No more Mr Nice Guy!
Managerrepents...
Tough managementworks!
Signal or Noise?
Reward!
Manager Repents
No more Mr Nice guy!
Tough Management works!
Signal or Noise?Signal or Noise?
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Velocity
ControlLimits
Signal or Noise?
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Velocity
ControlLimits
What to expect
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Velocity
Average
What to expect
What to expect
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Velocity
Average
Expectations...
10
20
30
40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Velocity
Expectations
Setting a TargetPe
rfor
man
ce
Time
✓
✗
Upper Control Limit
Lower Control Limit
Average
Setting a targetSetting a TargetPe
rfor
man
ce
Time
✓
✗
Upper Control Limit
Lower Control Limit
Average
“I recently asked a colleague [CIO] whether he would prefer to deliver a project somewhat late and overbudget, but rich with business benefits, or one that is on time and underbudget but of scant value to the business.
He thought it was a tough call, and then went for the on- time scenario. Delivering on time and within budget is part of his IT department’s performance metrics. Chasing after the elusive business value, over which he thought he had little control anyway, is not.”
Cutter Sr. Consultant Helen Pukszta
If you give a manager a numerical target, he’ll
make it, even if he has to destroy the company in
the process.
The Third Principle of the Theory of Variation
Work on the Causes of Variation,
Which are Always Found in the System
Majority of Performance is Down to the System
Majority of Performance is Down to the System
A bad system will defeat a good person
every time.
System
Individual
95%
5%
System Conditions Work Design Policies Measures Structure Roles Procedures Information Job skills Knowledge
Business Cases FundingPoor requirements
Inspection Compliance
The Fourth Principle of the Theory of Variation
Understanding Variation Tells you When
Something has Happened
Special Cause vs Common Cause
Special Cause vs Common Cause
10
20
30
40
50
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Velocity
Did we improve? Did we improve?
0
10
20
30
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Changeintroduced
after sprint 8
SPC for comparison SPC for comparison
Team A
Team B
Team C
Recommended