cmm g4 institutionalig process improvement
TRANSCRIPT
GTECH Corporation Copyright 20061Version 7 – Practical CMMI, April 2006
Institutionalizing Process Improvement: Is it a
Lottery?
Paul Morgan
Jurek Malecki
GTECH – Technology Process Group
April 24th, 2006
!
GTECH Corporation Copyright 20062Version 7 – Practical CMMI, April 2006
! Incorporated in 1980, headquartered in Rhode Island, USA with 5,300
employees worldwide in more than 50 countries and $1.25 billion in total
revenue in FY 2005.
! The leading provider to the world’s lottery industry with market share of more
than 70% and more than 438,000 point-of-sale devices linked to GTECH central
systems.
! Handles more transactions a year than all of the leading credit card companies
combined (50 billion transactions in FY04).
! In 2003 and 2005 software development organizations located in Austin,
Chennai, and Warsaw were formally assessed at CMM Level 3. Updated
CMMI/SW Level 4 processes were deployed in 2005.
! In 2006 it was announced that Lottomatica S.p.A. (Milan: LTO) and GTECH
have entered into an agreement pursuant to which Lottomatica will acquire
GTECH.
GTECH Corporate Profile
GTECH Corporation Copyright 20063Version 7 – Practical CMMI, April 2006
0.00
10.00
20.00
30.00
40.00
50.00
60.00
ATC CTC WTC Global
2004
2005
Defect Density(Defects Found per 100 GFP)
0%
20%
40%
60%
80%
100%
2000 2001 2002 2003 2004 2005
DRE
Std Dev
Defect Removal Efficiency
Over the course of its process improvement journey
GTECH has enjoyed:
" 60% Increase in requirements stability
" 55% Improvement in defect removal efficiency
(DRE)
" 40% Reduction in rework
" 80% increase in cost and schedule estimation
accuracy and predictability
0%
10%
20%
30%
40%
50%
60%
70%
80%
Inspection & Unit Test System Test CAT
Defe
ct D
ete
ction
2000
2005
Defect Detection Profile
Process Improvement Journey: Tangible Results
GTECH Corporation Copyright 20064Version 7 – Practical CMMI, April 2006
! The ingrained way of doing business that an organization follows routinely as
part of its corporate culture.
CMU/SEI-2002-TR-012 Capability Maturity Model® Integration (CMMIsm), Version
1.1 (Staged Representation). Appendix C Glossary, page 579
Institutionalization: CMMI Viewpoint
! In CMMI appraisals, institutionalization is judged by achievement of generic goals at the
appropriate capability level / maturity level.ARC Definition
Level 3 to 5
5.7%
Level 3 to 4
3.1%
Level 2 to 5
2.2%
Level 2 to 4
3.6%
Level 2 to 3
27.7%
Level 1 to 5
0.7%
Level 1 to 4
2.0%
Level 1 to 3
11.4%
Level 1 to 2
21.0%
Moved Down
2.9%Level 4 to 5
7.0%
Reappraisals – Change in Maturity LevelSEI Process Maturity Profile
September 2005
GTECH Corporation Copyright 20065Version 7 – Practical CMMI, April 2006
Is All What It Seems?: Process Stability and Capability
BatchLiveDate
Prod
uctiv
ity(G
FPs/
100 H
r s) _
1
1-Sep-20051-Jul-20051-May-20051-Mar-20051-Jan-20051-Nov-20041-Sep-2004
50
40
30
20
10
0
27
17
7
18-Jan-2005
Austin
Benelux
Chennai
Warsaw
Hub
Scatterplot of Productivity(GFPs/100 Hrs)_1 vs BatchLiveDate
Process has
become Stable
and Capable
Actual Project Size (GFPs)
Total Project Effort (100 Hrs)
Process is Not
Stable and Not
Capable
BatchLiveDate
ReqV
ol
1-Sep-051-Aug-051-Jul-051-Jun-051-May-051-Apr-051-Mar-05
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0.3
0.45
Austin
Chennai
Warsaw
Hub
Scatterplot of ReqVol vs BatchLiveDate
Total Number of Requirements Changed, Added, Deleted
Original Number of Requirements
GTECH Corporation Copyright 20066Version 7 – Practical CMMI, April 2006
! Conscious Competence Learning Model
‘There are known knowns. These are things we know that we know. There are known unknowns. That is to say,there are things that we know we don't know. But there are also unknown unknowns. There are things we don'tknow we don't know.’
- Donald Rumsfeld
! Mental Models
‘The Map is not the territory’
- Alford Korzybski
! Theory of Actions
! Single and Double Loop Learning
! Systems Thinking
‘Every man takes the limits of his own field of vision for the limits of the world ’
- Arthur Schopenhauer
! Systems Mapping
! Leverage Points
Institutionalization: Alternative Viewpoints
GTECH Corporation Copyright 20067Version 7 – Practical CMMI, April 2006
‘There are known knowns. These
are things we know that we
know….’
Skills acquired to overcome
the weakness but needs full
focus or supervision to
undertake activity.
‘……..……’
Skill becomes second nature
and can be successfully
accomplished without
conscious awareness.
‘…there are also unknown
unknowns. There are things we
don't know we don't know.’
No awareness that a
weakness exists or of the
need to develop new skills.
‘There are known unknowns. That
is to say, there are things that we
know we don't know….’
Awareness that an
improvement opportunity
exists and that new skills
need to be acquired.
Unconscious
Conscious
Competence Incompetence
Blissfully unaware that
anything is wrong or that
tying her shoelaces has any
benefits
After falling over a few times
recognises that shoelaces
should be tied but does not
know how to do it.
Can manage to tie them
perfectly, but needs to
concentrate each time on the
process.
Is able to tie her laces
without consciously paying
any attention to the process.
Conscious Competence Learning Matrix
GTECH Corporation Copyright 20068Version 7 – Practical CMMI, April 2006
Unconscious
Conscious
Competence Incompetence
Applicability to Process Improvement
Training
Redesign
Processes
Organizational
Appraisal
Mentoring Quality
Assurance
Unawareness
Discovery
Institutionalization
Repetition
GTECH Corporation Copyright 20069Version 7 – Practical CMMI, April 2006
Mental Models: The Theory of Actions
Espoused Theory
! The world view and values people
and organizations believe their
behaviour is based upon.
! Derived from politically correct
values an individual or organisation
wants others to accept as its guiding
principles, governance, or social
responsibility.
Theory-in-Use
! The world view and values implied
by the behaviour of a person or
organization.
! Determined by the unconscious
‘maps’ or ‘ mental models’ that filter
feedback and reinforce actions that
conform with the model and sabotage
those in conflict with it.
Maps TerritoryActions
GTECH Corporation Copyright 200610Version 7 – Practical CMMI, April 2006
Size allows us to better
estimate and control ourprojects.
The SEPG knowwhat they’re doing
and always make
the correct
decisions.
Size estimation is a
requirement of CMMI so I
do it as part of my job.
This sizing method is just
overhead and adds novalue. I only do it for a
quiet life and to keep
standards compliance off
my back.
Espoused Theory
Theory in Use(2)
(3)
Institutionalization: Theory of Actions and Learning
This size estimation is a waste
of time and adds no value toour existing way of working.
Next they’ll be
measuring
productivity and I’llrun the risk of
loosing my
reputation within the
company.
(1)
Mental Map Action Strategy Consequences
Single Loop
Double Loop
GTECH Corporation Copyright 200611Version 7 – Practical CMMI, April 2006
System Mapping: Process Improvement
Performance
Driver
Actual
Performance
Gap
Corrective
Action
Desired
State
Goal Seeking
Loop
Employee Motivation
Process
Capability
CMMI
Compliance
Training
Successful
Process Initiative
s
s
s
o
Compliance
Checks
Project Bonus
Shared Vision
s
s
o
“S” means an influence in
the “Same” direction
“O” means an influence in
the “Opposite” direction
Process
Institutionalization
NoBonus
NoTargets
o
GTECH Corporation Copyright 200612Version 7 – Practical CMMI, April 2006
Size Estimation: So What Went Right?
BatchLiveDate
Prod
ucti
vity
(GFP
s/10
0 H
rs)_
1
1-Sep-20051-Jul-20051-May-20051-Mar-20051-Jan-20051-Nov-20041-Sep-2004
50
40
30
20
10
0
27
17
7
18-Jan-2005
Austin
Benelux
Chennai
Warsaw
Hub
Scatterplot of Productivity(GFPs/100 Hrs)_1 vs BatchLiveDate
Process has
become Stable
and Capable
GTECH Corporation Copyright 200613Version 7 – Practical CMMI, April 2006
June 2002: Introduction of Size Estimation Methodology
Project
Requirements
System
Architecture
Complexity
rating
! June 2002 GTECH
Function Points
(GFP) introduced.
! Excel based tool.
! Estimation
responsibility of
project manager
but typically
delegated to design
personnel.
GTECH Corporation Copyright 200614Version 7 – Practical CMMI, April 2006
Estimated DRE: 95%Estimated DRE: 95%
Planning Requirements
Analysis
Design Code & Unit
Test
Integration
Test
System Test Customer
Test
Num
ber
of D
efe
cts
Dete
cte
d
Plan
Actual
Low
High
November 2004: Standard Project Process (SPP) Version 2.2
! Version 2.2 of SPP released in
November 2004 and contained for
the first time size based estimation
utilities.
Defect
Profile
Effort
Estimation
GTECH Corporation Copyright 200615Version 7 – Practical CMMI, April 2006
Employee Motivation
Process
Capability
CMMI
Compliance
Training
Successful
Process Initiative
s
s
s
o
Compliance
Checks
Project Bonus
Project Peer
Pressure
s
s
o
“S” means an influence in
the “Same” direction
“O” means an influence in
the “Opposite” direction
NoBonus
NoTargets
o
Process
Institutionalization
System Mapping: Sizing Process
Quantitative
Project Management
sS
GTECH Corporation Copyright 200616Version 7 – Practical CMMI, April 2006
! The concept of software size estimation is not intuitive for most technical personnel.
! Individuals, teams, and organizations operate within the unconscious incompetence zone of the learningmatrix.
! Training is ineffective in this situation. What is required is recognition of the relevance of the skill areabeing taught, and an awareness that a skill deficiency exists.
! Even where such conditions existed unsupportive mental models were impeding the institutionalization of thesizing methodology.
! Size estimation was perceived as bureaucratic and not seen as necessary by ‘experienced’ technicalpersonnel.
! The new method challenged the standing of senior technical personnel who traditionally providedestimates based upon ‘expert judgment’.
! Sizing was perceived as a management tool to enable productivity comparisons and to enforceaccountability.
! At best sizing was perceived to produce data for ‘post mortem’ investigations by which time practitionerswere already working on their next project.
! Introduction of the size based utilities produced dramatic results because:
! Benefits of size estimation became visible to all and surfaced unconscious incompetence.
! Size estimation was reframed from bureaucratic overhead to a professionally justified element of thetechnical engineering function.
! The new method was now seen to be complementary to ‘expert judgement’ estimation and not a threat.
! New utilities made ‘real time’ benefits visible at the individual, team, and organizational level.
! Productivity was measured and reported was seen to be non-attributable to individuals or teams.
Size Estimation – So What Went Right?
GTECH Corporation Copyright 200617Version 7 – Practical CMMI, April 2006
! When implementing process change select methods appropriate to the position of the
people, teams, and organization on the learning matrix.
! People and organizations have mental models which drive behavior. These models are
usually unconscious and self-protect against change.
! To institutionalize effective and sustainable change the mental models need to be updated
to drive the required behavior.
! Change agents should practice a systemic approach to problem solving wherein a
organization is analyzed as a whole rather than focusing in on individual component parts.
! Consideration should be given that cause and effect tend not to be closely related in time
and space and are often counter-intuitive.
! Small changes can produce big results via the identification of leverage points.
! Even expert change agents have areas of unconscious incompetence and are constrained by
the blind-spots in their mental models.
Take Home Thoughts
GTECH Corporation Copyright 200618Version 7 – Practical CMMI, April 2006
Thank You
Any Questions
Contact Information: [email protected]
GTECH website: http://www.gtech.com
GTECH Corporation Copyright 200619Version 7 – Practical CMMI, April 2006
Software Process Improvement and Dynamics Modelling
• Tarek Abdel-Hamid/Stuart E. Madnick: SOFTWARE PROCESS DYNAMICS, AN INTEGRATEDAPPRAOCH, Prentice Hall Software Series
• John D. Sterman: BUSINESS DYNAMICS, Systems Thinking and Modelling for a Complex World, IrwinMcGraw-Hill
• Gerald M. Weinberg: QUALITY SOFTWARE MANAGEMENT Volume 1: SYSTEMS THINKING;Dorset House Publishing
• Gerald M. Weinberg: QUALITY SOFTWARE MANAGEMENT Volume 3: CONGRUENT ACTION;Dorset House Publishing
• Gerald M. Weinberg: QUALITY SOFTWARE MANAGEMENT Volume 4: ANTICIPATING CHANGE;Dorset House Publishing
Systems Thinking
• Peter M. Senge: THE FIFTH DISCIPLINE, The Art & Practice of the Learning Organisation, RandomHouse
• Peter Senge (an others): THE DANCE OF CHANGE, The Challenges of Sustaining Momentum inLearning Organisations; Nicholas Brealey Publishing
• Peter Senge (and others): THE FIFTH DISCIPLINE FIELDBOOK, Nicholas Brealey Publishing
• Peter Senge (and others): PRESENCE, Exploring Profound Change in People, Organisations and Society;Nicholas Brealey Publishing
Sources and Further Reading (Part One)
GTECH Corporation Copyright 200620Version 7 – Practical CMMI, April 2006
Mental Maps and Organizational Change
• Chris Argyris: ON ORGANISATIONAL LEARNING, Second Edition, Blackwell Publishing
• Chris Argyris: KNOWLEDGE FOR ACTION: A Guide to Overcoming Barriers to Organisational Change;Jossey-Bass Publushers
• Chris Argyris: REASONS AND RATIONALISATION, The Limits to Organisational Knowledge; OxfordUniversity Press
Organizational Structures and Aspects of Culture
• Gerry Johnson, Kevan Scholes, Richard Whittington: EXPLORING CORPORATE STRATEGY, Text andCases; Prentice Hall, Financial Times, Seventh Edition
• David Bucanan, Andrzej Huczynski: ORGANISATIONAL BEHAVIOUR, Third Edition, Prentice Hall
Sources and Further Reading (Part Two)