Cmmi 26 ago_2009_

© 2008 Carnegie Mellon University
®
CMMI is registered in the U.S. Patent and Trademark Office by
Carnegie Mellon University.
CMMI®
Update
August 26, 2009
Mike Phillips
Software Engineering Institute
Carnegie Mellon University

2
CMMI Update
Phillips, Aug 26, 2009
Strategic Technological
Structural
Human/
Cultural
Managerial
Organizational
System
Input-output flow of materials, energy, information
Inputs:
Human,
Financial,
Technological,
Material,
Resources
Outputs:
Products,
Services
Organizations Are Complex Systems
Adapted from Kast and Rosenzweig, 1972.

3
CMMI Update
“M” is for Model
Models are
simplified views
of the real world.
CMM
THE REAL WORLD
Process
descriptions,
models, and
instantiations
are below the
level of detail
of the CMMs.
[Systems
Engineering]
Marketing
[Integrated product teams]
Technology
Organizational
culture
[People issues]
Maturity Levels
Process Areas
Practices
Process
Descriptions
“All models are wrong; some models are useful.” George Box

4
CMMI Update
How Do You Want to Work?
• Random motion – lots of energy, not
much progress
• No teamwork – individual effort
• Frequent conflict
• You never know where you’ll end up
• Directed motion – every step brings
you closer to the goal
• Coordinated efforts
• Cooperation
• Predictable results
Processes can make the difference!

5
CMMI Update
SEI’s IDEAL Approach
Diagnosing Establishing
Acting
Learning
Propose
Future
Actions
Analyze
and
Validate
Pilot/Test
Solution
Create
Solution
Develop
ApproachSet Priorities
Develop
Recommendations
Characterize
Current
and Desired
States
Charter
Infrastructure
Build
Sponsorship
Stimulus for
Change
Set Context
Implement
Solution
Refine
Solution
Plan Actions
Initiating

6
CMMI Update
SCAMPI Calendar Year-end 2008 Statistics
(Cumulative) as of 3-31-09
CYE07 CYE08 Delta % Increase
# Appraisals Performed 3,113 4,134 1,021 32.8%
# Unique Organizations Appraised 2,674 3,446 772 28.9%
# Unique Participating Companies 1,882 2,544 672 35.7%
# Re-appraised Organizations 361 564 203 56.2%
# Unique Projects 14,620 21,141 6,521 44.6%

7
CMMI Update
CMMI Transition Status
Reported to the SEI as of 7-31-09
Training
Introduction to CMMI - 105,442
Intermediate CMMI - 2,959
Understanding CMMI High Maturity Practices - 558
Introduction to CMMI V1.2 Supplement for ACQ - 939
Introduction to CMMI V1.2 Supplement for SVC (1 Day) - 668
Introduction to CMMI Services V1.2 (3 Day) - 46
Authorized / Certified
Introduction to CMMI V1.2 Instructors - 433
CMMI-ACQ V1.2 Instructors - 55
CMMI-SVC V1.2 Instructors - 69
SCAMPI V1.2 Lead Appraisers (195 Certified) - 520
SCAMPI V1.2 B & C Team Leaders - 518
SCAMPI V1.2 High Maturity Lead Appraisers (All Certified) - 149
CMMI-ACQ V1.2 Lead Appraisers (16 Certified)
CMMI-SVC V1.2 Lead Appraisers (All Certified)
- 56
- 52

8
CMMI Update
CMMI, V1.2 Foreign Language Translation Status
Reported to the SEI
CMMI-DEV, V1.2
Language Status
Chinese (Simplified) Underway, to be completed in 2009
Chinese (Traditional) Completed December 2007
French Completed August 2008
German Completed in April 2009
Japanese Completed August 2007. Intro course translated
October 2007
Portuguese To be completed in November 2009
Spanish Completed in June 2009
CMMI-ACQ, V1.2
Language Status
Chinese (Traditional) Completed April 2009

9
CMMI Update
Number of v1.1/v1.2 Introduction to CMMI Attendees by Year
and Quarter as of 7-31-09
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
5500
Q
1/0
4
Q
2/0
4
Q
3/0
4
Q
4/0
4
Q
1/0
5
Q
2/0
5
Q
3/0
5
Q
4/0
5
Q
1/0
6
Q
2/0
6
Q
3/0
6
Q
4/0
6
Q
1/0
7
Q
2/0
7
Q
3/0
7
Q
4/0
7
Q
1/0
8
Q
2/0
8
Q
3/0
8
Q
4/0
8
Q
1/0
9
Q
2/0
9
Q
3/0
9

10
CMMI Update
SCAMPI v1.1/v1.2 Class A Appraisals Conducted by Quarter
Reported as of 7-31-09
0
50
100
150
200
250
300
350
400

11
CMMI Update
CMMI Adoption Has Been Broad
25 countries with 10 or more
appraisals (Aug 06 -> Jul 08):
• USA 598 -> 1034
• China 158 -> 465
• India 177 -> 323
• Japan 155 -> 220
• France 65 -> 112
• Korea (ROK) 56 -> 107
• Taiwan 31 -> 88
• Brazil 39 -> 79
• Spain 25 -> 75
• U.K. 42 -> 71
• Germany 28 -> 51
• Argentina 15 -> 47
• Canada 18 -> 43
• Malaysia 15 -> 42
• Mexico <10 -> 39
• Australia 23 -> 29
• Egypt 10 -> 27
• Chile <10 -> 20
• Philippines 14 -> 20
• Colombia <10 -> 18
• and Italy, Israel, Singapore,
Hong Kong, and Pakistan
http://www.sei.cmu.edu/appraisal-program/profile/profile.html
Statistical analyses by the presenter.
Business Services
40.8%
Engineering & Management
Services
13.3% Other Services
0.8%
Mining
0.2%
Wholesale Trade
1.0%
Transportation,
Communication, Electric, Gas
and Sanitary Services
3.2%
Retail Trade
0.4%
Finance, Insurance and Real
Estate
5.9%
Public Administration
(Including Defense)
10.1%
Fabricated Metal Products
0.4%
Primary Metal Industries
1.0%
Industrial Machinery And
Equipment
2.2%
Instruments And Related
Products
5.9%
Electronic & Other Electric
Equipment
5.1%
Transportation Equipment
9.1%Health Services
0.6%
Est’d 900,000+ work in
orgs that have had a
SCAMPI A appraisal.
• SCAMPI A reports from 60 countries
• 72% of adopters are commercial orgs
• 2/3 Services; 1/5 Manufacturing
• Approx. 70% of adopters in US are
contractors for military/gov’t or are gov’t
25 or fewer
13.6%
26 to 50
15.6%
51 to 75
12.8%
76 to 100
8.9%
101 to 200
19.8%
201 to 300
9.0%
301 to 500
7.4%
501 to 1000
6.8%
1001 to 2000
3.7%
2000+
2.5%
201 to 2000+
29% 1 to 100
51%
Manufacturing
19%
Services
66%

12
CMMI Update
Countries where Appraisals have been Performed
and Reported to the SEI
Argentina Australia Austria Bahrain Bangladesh Belarus Belgium Brazil
Bulgaria Canada Chile China Colombia Costa Rica Czech Republic Denmark
Dominican Republic Egypt Finland France Germany Greece Hong Kong Hungary
India Indonesia Ireland Israel Italy Japan Korea, Republic Of Latvia
Lithuania Luxembourg Malaysia Mauritius Mexico Morocco Nepal Netherlands
New Zealand Norway Pakistan Panama Peru Philippines Poland Portugal
Romania Russia Saudi Arabia Singapore Slovakia South Africa Spain Sri Lanka
Sweden Switzerland Taiwan Thailand Turkey Ukraine United Arab Emirates United Kingdom
United States Uruguay Viet Nam

13
CMMI Update
Number of Appraisals and Maturity Levels
Reported to the SEI by Country
Country
Number of
Appraisals
Maturity
Level 1
Reported
Maturity
Level 2
Reported
Maturity
Level 3
Reported
Maturity
Level 4
Reported
Maturity
Level 5
Reported Country
Number of
Appraisals
Maturity
Level 1
Reported
Maturity
Level 2
Reported
Maturity
Level 3
Reported
Maturity
Level 4
Reported
Maturity
Level 5
Reported
Argentina 64 45 12 2 3 Malaysia 56 20 31 5
Australia 32 1 7 5 2 4 Mauritius 10 or fewer
Austria 10 or fewer Mexico 57 24 25 3 4
Bahrain 10 or fewer Morocco 10 or fewer
Bangladesh 10 or fewer Nepal 10 or fewer
Belarus 10 or fewer Netherlands 10 or fewer
Belgium 10 or fewer New Zealand 10 or fewer
Brazil 106 1 50 42 1 9 Norway 10 or fewer
Bulgaria 10 or fewer Pakistan 25 1 18 4 1
Canada 51 1 12 22 5 3 Panama 10 or fewer
Chile 30 17 10 2 Peru 10 or fewer
China 745 1 117 540 27 41 Philippines 21 2 11 7
Colombia 22 7 11 1 2 Poland 10 or fewer
Costa Rica 10 or fewer Portugal 10 or fewer
Czech Republic 10 or fewer Romania 10 or fewer
Denmark 10 or fewer Russia 10 or fewer
Dominican Republic 10 or fewer Saudi Arabia 10 or fewer
Egypt 34 1 17 11 2 2 Singapore 19 3 10 1 4
Finland 10 or fewer Slovakia 10 or fewer
France 141 4 81 45 1 2 South Africa 10 or fewer
Germany 64 9 32 11 1 1 Spain 105 1 60 35 2 4
Greece 10 or fewer Sri Lanka 10 or fewer
Hong Kong 18 2 11 5 Sweden 10 or fewer
Hungary 10 or fewer Switzerland 10 or fewer
India 409 14 191 24 166 Taiwan 117 1 74 38 2
Indonesia 10 or fewer Thailand 27 12 13 1
Ireland 10 or fewer Turkey 14 12 2
Israel 17 3 10 2 Ukraine 10 or fewer
Italy 31 14 14 United Arab Emirates 10 or fewer
Japan 267 17 75 121 13 16 United Kingdom 93 3 42 30 1 3
Korea, Republic Of 138 1 47 61 13 7 United States 1272 27 448 462 21 124
Latvia 10 or fewer Uruguay 10 or fewer
Lithuania 10 or fewer Viet Nam 12 9 1 2
Luxembourg 10 or fewer

Beyond CMMI V1.2...

15
CMMI Update
3 Complementary “Constellations”
CMMI-SVC
CMMI-DEV
CMMI-Services
provides guidance for
those providing
services within
organizations and to
external customers
CMMI-ACQ
CMMI-ACQ
provides
guidance to
enable
informed and
decisive
acquisition
leadership
CMMI-Dev
provides guidance
for measuring,
monitoring and
managing
development
processes
16 Core
Process Areas,
common to all

16
CMMI Update
CMMI-ACQ v1.2
Acquisition Category Process Areas
Acquisition
Requirements
Development
Solicitation &
Supplier
Agreement
Development
Agreement
Management
Acquisition
Technical
Management
Acquisition
Validation
Acquisition
Verification
CMMI Model
Framework
(CMF)
16 Project,
Organizational,
and Support
Process Areas

17
CMMI Update
Visibility into the Team’s Capability
Acquisition
Planning
RFP
Prep.
Solicita-
tion
Source
Selection
System
Acceptance
Program Leadership
Insight / Oversight
Transition
Plan Design
Integrate
& Test
Develop Deliver
CMMI for Development
CMMI for Acquisition
Operational
Need
Developer
Acquirer

18
CMMI Update
CMMI-SVC v0.5(+)
Capacity
and
Availability
Managemen
t
Service
Continuity
Service
System
Development
Strategic
Service
Managemen
t
Incident
Resolution
&
Prevention
Service
Delivery
Service
System
Transition
16 Core
Process Areas
and 1 shared
PA (SAM)
PA Addition

19
CMMI Update
People Capability Maturity Model: Primary
Objective
Performance
Process
capability
Workforce
capability
enables improves
People CMMPeople CMM CMMI-DEV,
-ACQ, -SVC
CMMI-DEV,
-ACQ, -SVC
The primary objective of:
a CMM is to improve the capability of an organization’s processes.
CMMI (DEV, ACQ, SVC) is to improve the capability of an
organizations processes within specific domains.
the People CMM is to improve the capability of an organization’s
workforce through enhanced management and human capital
processes.
(People CMM defines capability as the level of knowledge, skills, and
process abilities available within each workforce competency of the
organization to build its products or deliver its services.)

20
CMMI Update
CMMI
Planned Elements – Multi-Model
•Improving the
interfaces is of
interest to both
government and
industry….

21
CMMI Update
CMMI
Planned Elements – Multi-model2
•Multiple models complicate process improvement –
but make it much more powerful by addressing
specific needs in various environments….

22
CMMI Update
Planned Sequence of Models
SA-CMMSA-CMM
GM IT SourcingGM IT Sourcing
CMMI-DEV V1.2CMMI-DEV V1.2
CMMI-ACQCMMI-ACQ
CMMI-SVCCMMI-SVC
CMMI V1.1CMMI V1.1
CMMI-AMCMMI-AM
CMMI V1.3CMMI V1.3
March
2009
2010
41

23
CMMI Update
CMMI Version 1.3
Product Suite:
What May Change
Mike Konrad
Mike Phillips
Rusty Young
Will Hayes
Aug 26, 2009

24
CMMI Update
CMMI Product Suite, Version 1.31
• “Version 1.3 will focus upon but not be limited to:
• High Maturity
• More effective GPs
• Appraisal efficiency
• Consolidation across Constellations”
• (per criteria documented by the CMMI Steering Group)
• Version 1.3 will be Change Request (CR) driven.
• Events such as “Beyond Version 1.2,” “Future Directions,” and this webinar are for
information sharing and dialogue

25
CMMI Update
CMMI Product Suite, Version 1.32
• We would like to present some ideas around certain topics.
• For selected topics, we will briefly discuss:
• Our initial ideas on the topic
• Pros/Cons
• For selected ideas, we may poll you for your preference
• Candidate questions (we’ll ask as time permits)
are in green font
• The poll will give us insight into the preferences of the community.

26
CMMI Update
High Maturity Topics
• Overall goal: improve clarity of HM practices
• Introduce a lower-ML version of CAR
• Show Practice Evolution through Amplifications

27
CMMI Update
Overall Goal: Improve Clarity of HM Practices
• Problem statement:
• HM practices are currently unclear, leading to a variety of interpretations
• The goal in a nutshell:
• All CMMI users have a common understanding of the HM Practices
• Involves:
• Clarifying connection between statistical management of subprocesses and project
management
• Increase depth and clarity of CAR and OID practices
• Improving/clarifying terminology
• Aligning required, expected, and informative
— Also, clarify the role of informative material in the model
• Adding additional notes and examples describing typical “how to’s”

28
CMMI Update
Introduce a lower-ML version of CAR
• Causal Analysis is viewed as valuable at any ML
— Use of OSSP, and PDP does ‘stabilize’ performance to a degree
— ML3 basis can assure some uniformity of impact
— May not be able to distinguish special from common causes of variation
• Create a lower ML version of CAR (e.g., as an SP[s] in IPM).
• Involves:
• Strengthening CAR at ML 5
• Introducing practices (e.g., IPM) addressing causal analysis at ML 3

29
CMMI Update
Show Practice Evolution through Amplifications
• Some see ML 2-3 practices as static – not evolving as ML increases
• Make it easier for an organization to interpret the practices as the organization matures.
• Involves:
• Reviewing lower-ML practices for instances where an amplification would clarify how the
practice “matures” at levels 4 and 5 (e.g., PP)
• Inserting clearly-delimited amplifications in lower-ML PAs

30
CMMI Update
More Effective GPs Topics
• Overall goal: simplify the GPs
• Reducing the GG 2 GPs to a Few
• Clarifying role of measurement in GP 2.8
• Improving the Clarity of GP 3.2
• Eliminating Generic Practices at CL 4-5

31
CMMI Update
Overall Goal: Simplify the GPs
• Problem Statement:
• GPs have a multiplier effect on # of items that need to be addressed
— GPs greatly affect the # of artifacts needed for an appraisal
— If a GP is misunderstood, every PA is affected
• Redundancy: Almost all GPs have associated PAs
• Simplify the GPs
• Involves:
• Reviewing the need for GG 1, GG 4-5, and their GPs
• Reducing the # of GPs at CL 2 (which are more important?)
• Rewording problematic GPs (per Pat O’Toole’s ATLAS surveys)
• Clarifying that GPs can be implemented at single or multiple organizational or project
levels, or both, depending on the PA.

32
CMMI Update
Reducing the GG 2 GPs to a Few (1)
• A proposal to reduce the GPs under GG 2 to just a few was presented at a webinar a
few months ago:
• GP 2.1 Policy
• GP 2.2 Plan the Process
• GP 2.8 Monitor and Control the Process
(revised to incorporate GP 2.10 Higher-level management)
• Webinar feedback suggests the above may be too few.
• Would you:
A. Add 2.3 Resources?
B. Add 2.9 Adherence?
C. Keep 2.10 separate?
D. Not add any other generic practices?
E. Keep all the generic practices?

33
CMMI Update
Reducing the GPs to a few (2)
• Instead of reducing the GPs to a few, would you
A. Add content and examples to the model clarifying the intent of the GPs and
describing how to address the overlap between GPs and PAs?
B. Change the appraisal method with respect to artifacts required for GPs?
C. Still reduce the GPs

34
CMMI Update
Clarifying Role of Measurement in GP 2.8
• Measuring the process should be seen as a means (as opposed to the
means) to monitoring and controlling the process
• GP 2.8 was never intended to require a
measure for every PA
• Informative material should be updated to clarify
this
• But if no measures are ever involved in
monitoring and controlling the process for any
PA, this should raise a “red flag”

35
CMMI Update
Improving the Clarity of GP 3.2
• Propose rewording GP 3.2 from:
• “Collect work products, measures, measurement results, and improvement information
derived from planning and performing the process to support the future use and
improvement of the organization’s processes and process assets”
• to:
• “Collect information derived from planning and performing the process to support the
future use and improvement of the organization’s processes.”
• Replace “information” with:
A. “information”
B. “measurements”
C. “measurements and other information”

36
CMMI Update
Eliminating Generic Practices at CL 4-5
• CL 4-5 GPs inadequately cover PAs with which they are associated
• Used in only a minority of appraisals
— Not necessary for Staged representation
• Encourage suboptimization
• Remove the CL 4-5 GPs until a broad and shared understanding of the associated
PAs at ML4-5 is achieved.
• Involves:
• Simplifying insitutionalization discussion for both representations (e.g., equivalent
staging)
• Should the CL 4-5 GPs be eliminated from V1.3?
• A. Yes. B. Don’t care. C. No

37
CMMI Update
Appraisal Efficiency
• Possible changes to the model to improve efficiency include:
• Simplifying # of GPs (see earlier slides)
• Resolve practice ambiguities
— E.g., PP SP 1.1, “Establish … WBS to estimate … scope …”
• Make applicability of SPs more “solid” (e.g., evaluating supplier processes)
• Refinements to SCAMPI itself include:
• Using directed sampling (during Discovery)
— Constructing an appropriate indicator set (rotating, random)
— Readiness review (ensuring indicators have been identified)
• Using Inverted PIIDs (as an option, not a requirement)?
— Documents tagged with all the SPs and GPs they map to

38
CMMI Update
Consolidation Across Constellations1
• Overall goal:
• Achieve balance between:
— “Naturalness” of model content within each area of interest
— Commonality of model content across constellations
• Help ensure CMMI Models are usable in combination
• Provide clarity as to what is specific to an area of interest vs. common
• Reference:
• CMMI-DEV V1.2, CMMI-ACQ V1.2, forthcoming CMMI-SVC V1.2

39
CMMI Update
• Will your org’n be using multiple CMMI models?
A. No
B. DEV+ACQ
C. DEV+SVC
D. SVC+ACQ
E. All 3

40
CMMI Update
• Covering Integrated/Self-Directed Teams in CMMI:
• In CMMI-DEV, this topic is optional but at goal level (OPD SG 2, IPM SG 3)
• In CMMI-ACQ, this topic is expected (OPD SP 1.7, IPM SP 1.6)
• People CMM has considerable material on teams
• Definitions
• IPT = Integrated Product Team (all disciplines having a significant role in product
lifecycle are represented on IPT)
• Self-directed team = A team that establishes its own plans and processes; measures,
analyzes, and improves its performance; and adjusts in response to change
• Select type of team (A) and how included (B)
• A1. IPT B1. Optional as in DEV
• A2. Self-directed B2. Expected as in ACQ, SVC
• A3. Both

41
CMMI Update
• Some reviewers of CMMI-SVC model drafts have proposed adding these PAs (with
appropriate wording changes) to DEV and ACQ:
• Capacity and Availability Management (CAM)
— The purpose of CAM is to ensure effective service system performance and ensure
that resources are provided and used effectively to support service requirements.
• Service Continuity (SCON)
— The purpose of SCON is to establish and maintain plans to ensure continuity of
services during and following any significant disruption of normal operations.

42
CMMI Update
Miscellaneous Topics
• Multiple model implementations
• Agile methods
• “Lean” the Model
• Additional Guidance
• Repackaging the model

43
CMMI Update
Multiple Model Implementations
• Enable organizations to implement multiple models, frameworks, etc.
• Reference:
• Jeannine Siviy’s and Patrick Kirwin’s SEI Webinar, “Process Improvement in Multi-
Model Environments (PrIME),“ July 2008
• Involves:
• Ensuring CMMI constellations have common Architecture & Glossary
• A methodology to efficiently trace to multiple frameworks

44
CMMI Update
Agile Methods
• Enable CMMI and Agile users to make synergistic use of both
• Reference:
• Hillel Glazer, Jeff Dalton, David Anderson, Mike Konrad, and Sandy Shrum, “CMMI®
or Agile: Why Not Embrace Both!” SEI Technical Note CMU/SEI-2008-TN-003.
November 2008.
• Involves:
• Inserting notes that help adopters of Agile methods correctly interpret practices (e.g.,
evaluating alternative technical solutions [TS SG 1])

45
CMMI Update
“Lean” the Model
• Make the essence of the “underlying model” visible
• Reference:
• Results of Beyond V1.2 Workshops
• Involves:
• Identifying material that does not need to be in the model
• Moving this material to a Web page, Wiki, or separate technical note
• Making the external material easy to locate
• Eliminating/consolidating low-value practices
• Example:
• Move the GP Elaborations out of the model and into supplementary material

46
CMMI Update
Provide Additional Guidance
• Identify particular challenges (or opportunities) affecting the CMMI user community that
complicate interpretation and implementation of CMMI practices
• Make additional guidance, notes, and examples available (not necessarily within the
model)
• Should V 1.3 cover:
A. 6 Sigma, Lean
B. OPM3
C. People CMM
D. COTS, GOTS
E. System-of-system

47
CMMI Update
Repackaging the model
• Make the model easier to use
• Make the use across constellations cost-effective
• Clearly identify “normative” material
• Should V 1.3 cover:
A. Package all constellations in a single report/book and differentiate between
constellations via the use of appendices, amplifications, etc.?
B. Have a separate report/book for each constellation?
C. Separate the “normative” material (goals and practices) from the informative
material ?
D. Keep the current packaging of the model?

48
CMMI Update
Summary
• Four drivers for Version 1.3:
• Clarify high maturity,
• Simplify GPs,
• Increase appraisal efficiencies, and
• Improve commonality across the constellations.
• We appreciate your input and feedback!

49
CMMI Update
What Have We Missed?
•Now let’s chat….

Cmmi 26 ago_2009_

Recommandé

Recommandé

Contenu connexe

En vedette

En vedette (17)

Similaire à Cmmi 26 ago_2009_

Similaire à Cmmi 26 ago_2009_ (20)

Plus de JULIO GONZALEZ SANZ

Plus de JULIO GONZALEZ SANZ (20)

Cmmi 26 ago_2009_

Notes de l'éditeur