SCAMPI

Standard CMMI Appraisal Method for

Process Improvement

Chang-Hyun Jo
Department of Computer Science
California State University Fullerton
[email protected]
http://jo.ecs.fullerton.edu

Contents
„ Objectives
„ SCAMPI
„ Appraisals and Benchmarking
„ Appraisal Considerations
„ SCAMPI MDD
„ MDD Process
„ Finding and Rating
„ Document Appraisal Results
„ Appraisal Disclosure Statement (ADS)
„ SCAMPI Published Appraisal Results

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 2

1
 Many parts of this document are directly excerpted from the references listed.

Objectives
„ This document has been prepared mainly for
the students in my Software Engineering
class:
„ Who want to learn CMM/CMMI
„ Who want to understand and improve their
capability to develop software effectively
„ Who want to understand the key practices that
are part of effective processes for developing or
maintaining software, and to identify the key
practices that are needed to achieve the next
maturity level in the CMMI
„ Who want to identify the risks of having a
particular organization perform the work of a
contract
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 3

SCAMPI

Standard CMMI Appraisal Method

f P
for Process IImprovementt
(SCAMPISM)

2
Appraisals and Benchmarking
„ Many organizations find value in
benchmarking their progress (e.g.,
ascertaining maturity level scores, a
capability level profile) in process
improvement for both internal purposes and
with external customers and suppliers.
„ Process appraisals focus on identifying
improvement opportunities and
understanding the organization’s position
relative to the selected model or standard.

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 5

Appraisals and Benchmarking

„ Appraisal teams use CMMI to guide their
identification and prioritization of findings.
„ These findings, with guidance provided by
CMMI practices, are used to plan
improvements for the organization.
„ The appraisal principles for the CMMI Product
Suite remain the same as those used in
appraisals for other process-improvement
models.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 6

3
Appraisals and Benchmarking
„ Those principles are:
„ Senior management sponsorship
„ A focus on the organization’s business objectives
„ Confidentiality for interviewees
„ Use of a documented appraisal method
„ Use of a process reference model (CMMI) as a
base
„ A collaborative team approach
„ A focus on actions for process improvement

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 7

Appraisals and Benchmarking

„ The CMMI Product Suite provides a
rigorous appraisal method for
benchmarking that implements these
appraisal principles.
„ It is called the Standard CMMI Appraisal
pp
Method for Process Improvement
(SCAMPISM).

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 8

4
Appraisals and Benchmarking
„ For benchmarking against other organizations,
appraisals must ensure consistent ratings.
„ The achievement of a specific maturity level
or the satisfaction of a process area must
mean the same thing for different appraised
organizations.
„ Rules for ensuringg this consistencyy are
provided in the SCAMPI Method Definition
Document (MDD).

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 9

Appraisals and Benchmarking

„ The International Organization for
Standardization and International Electro-
technical Commission (ISO/IEC) Technical
Report 15504 is an emerging standard for SW
process assessment.
„ The CMMI models and SCAMPI were written to
support the conduct of appraisals that conform to
the 1998 version of Technical Report 15504.
„ Sponsors interested in performing an ISO/IEC
15504-conformant CMMI appraisal must provide
necessary information including a demonstration
of compatibility to support their needs.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 10

5
Appraisal Considerations
„ Choices that affect a CMMI
CMMI-based
based appraisal
include the following:
„ Establishing the appraisal scope, including the
organizational entity to be appraised, the CMMI
process areas to be investigated, and the maturity
level or capability level(s) to be appraised.
„ Selecting the appraisal method
„ Selecting the appraisal team members
„ Selecting appraisal participants from the appraisal
entities to be interviewed
„ Establishing appraisal outputs (e.g., ratings,
instantiation-specific findings)
„ Establishing appraisal constraints (e.g., time spent
on site) SCAMPI - Chang-Hyun Jo @ CSU Fullerton 11

SCAMPI MDD

Standard CMMI® Appraisal Method for

Process Improvement
(
(SCAMPI SM) A, Version 1.2:

Method Definition Document (MDD)

CMU/SEI-2006-HB-002

6
 Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI MDD
„ The Standard CMMI® Appraisal Method for
Process Improvement (SCAMPISM) A is
designed to provide benchmark quality
ratings relative to Capability Maturity Model®
Integration (CMMI) models.
„ It is applicable to a wide range of appraisal usage
modes including both internal process
modes,
improvement and external capability
determinations.
„ SCAMPI A satisfies all of the Appraisal
Requirements for CMMI (ARC) requirements for a
Class A appraisal method.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 13

SCAMPI MDD
„ The SCAMPI v1.2 Class A Method Definition
Document (MDD) describes the requirements,
activities, and practices associated with each
of the processes that compose the SCAMPI A
method.
„ It is intended to be one of the elements of the
infrastructure within which SCAMPI Lead
AppraisersSM conduct a SCAMPI A appraisal. appraisal
„ Precise listings of required practices, parameters,
and variation limits, as well as optional practices
and guidance for enacting the method, are
covered.
„ An overview of the method’s context, concepts,
and architecture is also provided.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 14

7
 Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI MDD
Table I-2 Part II Contents

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 15

SCAMPI MDD
„ People who will enact an appraisal
(appraisers) are expected to read the second
part of the document (MDD).
„ Appraisers need to know how to enact the method,
not just what the method is.
„ MDD’s Part II is divided into Process
Definitions, which are in turn divided into
A ti it Descriptions.
Activity D i ti
„ Each Activity Description delineates Required
Practices, Parameters and Limits, Optional
Practices, and Implementation Guidance.

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 16

8
 SCAMPI MDD
„ There are several processes contained in SCAMPI
A.
„ The processes (as defined) support a variety of
orderings and enactments to facilitate a variety of
usage modes for SCAMPI A.
„ The temporal flow, as well as the flow of inputs and
outputs among the processes, is described in the
Method Overview section.
„ The Process Definitions are not intended to provide a
start-to-finish view of SCAMPI A.
„ Instead, these sections provide detailed definitions of
processes and activities that are invoked according to
the appraisal plan created by the appraisal team
leader.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 17

Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

MDD

„ Table II-4
4
Process
Definition
Elements

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 18

9
 Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMI MDD
„ Table I-5:
I 5: Activity Description Elements

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 19

SCAMI Appraisal Classes

„ The SCAMPI family of appraisals includes
Class A, B, and C appraisal methods.
„ The Appraisal Requirements for CMMI (ARC)
describes a full benchmarking class of
appraisal as Class A.
„ Other CMMI-based appraisal methods might
pp p
be more appropriate for a given
g set of
sponsor needs, including self-assessments,
initial appraisals, quick-look or mini-appraisals,
incremental appraisals, and external
appraisals.

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 20

10
 SCAMI Appraisal Classes
„ SCAMPI A is the most rigorous method and
the only method that can result in a rating.
„ SCAMPI B provides options in model scope,
but the characterization of practices is fixed
to one scale and is performed on
implemented practices.
„ SCAMPI C provides a wide range of options,
including characterization of planned
approaches to process implementation
according to a scale defined by the user.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 21

SCAMI Appraisal Classes

„ Characteristics of Appraisal Classes

Characteristic Class A Class B Class C

Amount of objective evidence High Medium Low

Ratings generated Yes No No

Resource needs High Medium Low

Team size Large Medium Small

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 22

11
 CMMI Appraisals
„ SCAMPI incorporates best practices found successful
in the appraisal community, and is based on the
features of several legacy appraisal methods,
including:
„ CMM-Based Appraisal for Internal Process Improvement
(CBA IPI) V1.1 [Dunaway 96b]
„ Electronic Industries Alliance/Interim Standard (EIA/IS)
731.2 Appraisal Method [EIA 98b]
„ Software Capability Evaluation (SCESM) V3.0 Method
Description [Byrnes 96]
„ Software Development Capability Evaluation (SDCE) [AFMC
94]
„ FAA Appraisal Method (FAM) [Ibrahim 99]

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 23

Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI

„ Table II-8:
8:
Practice
Implementation
Indicator Types

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 24

12
 Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
„ An example of
how Practice
Implementation
Indicator (PII)
for each
indicator types
can be used in
verifying the
implementation
of a model
practice

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 25

Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
Table I-9: SCAMPI A Phase Summary:
Plan and Prepare for Appraisal

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 26

13
 Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
Table I-10: SCAMPI A Phase Summary:
Conduct Appraisal

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 27

Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
Table I-11: SCAMPI A Phase Summary:
Report Results

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 28

14
 Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
Figure I-3:
Process Flows, Plan and Prepare Processes

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 29

Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
Figure I-4:
Process Flows, Conduct Appraisal Processes

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 30

15
 Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
Figure I-5: Process Flows, Report Results Processes

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 31

Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

MDD
Process
„ MDD - SCAMPI - Phase 1
„ Process Definition 1.1
„ Process Activity 1.1.1, …
„ SCAMPI Phase Summary:
„ MDD Table I-9, I-10, I-11
„ Format
„ Table I-4 Process Definition Elements
„ Table I-5 Activity Description Elements

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 32

16
SCAMPI
MDD 2.6.1
Derive Findings and Rate Goals
„ The appraisal team shall derive final findings using
preliminary findings statements, feedback from
validation activities, and any additional objective
evidence collected as a result of the validation
activities.
„ The appraisal team shall rate each specific goal and
generic goal within the reference model scope of the
appraisal, based on the practice implementation
characterizations at the organizational unit level as
well as the aggregation of weaknesses associated
with that goal.
„ The appraisal team shall obtain appraisal team
consensus on the findings statements and ratings
generated for the organizational unit level.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 33

SCAMPI
MDD 2.6.1
Derive Findings and Rate Goals
„ When deriving final findings,
findings the aim is to
create goal-level statements that summarize
the gaps in practice implementation.
„ These statements must be abstracted to the
level of the organizational unit, and cannot
focus on individual projects or increments
(unless the tailoring option for project- or
increment-specific findings has been agreed
on during planning).

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 34

17
SCAMPI
MDD 2.6.1
Derive Findings and Rate Goals
„ A goal must be rated Not Rated if there are
any associated practices that are not
characterized at the organizational unit level
or that are characterized as Not Yet at the
organizational unit level.
„ A goal is rated Not Rated if the associated set
of objective evidence does not meet the
defined criteria for sufficient data coverage.

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 35

SCAMPI
MDD 2.6.1
Derive Findings and Rate Goals
„ The goal is rated Satisfied if and only if
„ all associated practices are characterized at the
organizational unit level as either Largely
Implemented or Fully Implemented, and
„ the aggregation of weaknesses associated with
the goal does not have a significant negative
impact on goal achievement.
„ For a goal to be rated as Unsatisfied, the
team must be able to describe how the set of
documented weaknesses (or single weakness)
led to this rating.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 36

18
 Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
MDD 2.4 Verify Objective Evidence
MDD 2.4.2 Characterize Implementation of Model Practices

„ Rules for characterizing

instantiation-level implementations
of practices
„ Consensus of at least a subset of
appraisal team members (mini-
team) is necessary for
instantiation-level characterizations.
„ The appraisal team shall
„ characterize for each instantiation
characterize, instantiation,
the extent to which appraisal
reference model practices are
implemented
„ aggregate practice implementation
characterization values from the
instantiation level to the
organizational unit level
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 37

Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
MDD 2.4 Verify Objective Evidence
MDD 2.4.2 Characterize Implementation of Model Practices

„ Rules for aggregating

instantiation-level
characterizations to
derive organizational unit-
level characterizations
„ Consensus of all members
of the appraisal team is
necessary for
organizational unit-level
characterizations.

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 38

19
SCAMPI
MDD 2.6.2b
Determine Satisfaction of Process Areas (PAs)
„ When using the Staged Representation of the
appraisal reference model, the team may
derive the satisfaction of PAs from the set of
goal satisfaction judgments.
„ The appraisal team shall rate the satisfaction
of each PA in the scope of the appraisal
based on the ratings of the goals within each
PA, if this rating option was selected during
planning.

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 39

SCAMPI
MDD 2.6.2b
Determine Satisfaction of Process Areas (PAs)
„ PAs must be assigned rating values of
Satisfied, Unsatisfied, Not Applicable, Out of
Scope, or Not Rated.
„ A PA is rated Satisfied if and only if all of its
specific and generic goals associated with a
given maturity level and below are rated
Satisfied
Satisfied.
„ If even one of the goals associated with a
given maturity level in a PA is rated
Unsatisfied, then the PA is rated Unsatisfied
for that maturity level and above.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 40

20
SCAMPI
MDD 2.6.2b
Determine Satisfaction of Process Areas (PAs)
„ When a PA is determined to be outside of the
organizational unit’s scope of work, the PA is
designated as Not Applicable and is not rated.
„ The identification of a PA as Not Applicable must
occur during the planning of the appraisal.
„ When an applicable PA is outside the scope of
the appraisal reference model used for the
appraisal the PA is designated as Out of
appraisal,
Scope and is not rated.
„ If even one of the goals in a PA is rated Not
Rated and none of the other goals are rated
Unsatisfied, then the PA is rated Not Rated.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 41

SCAMPI
MDD 2.6.2a
Determine Process Area CL
„ When using the Continuous Representation of
the appraisal reference model, the team may
make rating judgments about each PA (and
associated capability level) within the scope
of the appraisal.
„ Assigning capability level ratings is an optional
activity, selected at the discretion of the appraisal
sponsor and documented in the appraisal input.

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 42

21
 SCAMPI
MDD 2.6.2a
Determine Process Area CL
„ The appraisal team shall
„ rate the capability levels for each PA within the scope of
the appraisal, based on the highest level and all levels
below for which its specific goals and the generic goals
within the appraisal scope have been satisfied (if this rating
option was selected during planning).
„ if any goals for a PA are rated Not Rated, the PA will be
rated Not Rated and a capability level will not be assigned.
„ when a PA is determined to be outside of the
organizational unit’s scope of work, the PA is designated as
“not applicable” and is not rated, and a capability level will
not be assigned.
„ when an applicable PA is outside of the scope of the model
used for the appraisal, the PA is designated as “out of
scope” and is not rated, and a capability level will not be43
assigned. SCAMPI - Chang-Hyun Jo @ CSU Fullerton

Source: http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009

SCAMPI
MDD 2.6.2a
Determine Process Area CL

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 44

22
SCAMPI
MDD 2.6.3b
Determine Maturity Level
„ When using the Staged Representation,
Representation the
maturity level determined is the highest level
at which all PAs contained within the maturity
level, and within all lower maturity levels, are
rated as Satisfied or Not Applicable.
„ The single exception to this rule is that generic
goal 3 for applicable maturity level 2 PAs must
also be rated Satisfied for a maturity level rating
of 3 or higher to be determined.

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 45

SCAMPI
MDD 2.6.3b
Determine Maturity Level
„ When using the Continuous Representation,
the appraisal reference model provides for
equivalent staging, whereby a Capability
Profile can be used to derive an equivalent
maturity level rating.
„ A maturity level for a continuous representation is
achieved if the Capability Profile is at or above the
target profile for all PAs for that maturity level and
all lower maturity levels in the equivalent staging,
excepting those PAs that are designated as Not
Applicable.
„ The equivalence of particular Capability Profiles
and particular maturity levels is addressed in
Chapter 3 of the reference model.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 46

23
SCAMPI
MDD 2.6.4
Document Appraisal Results
„ The appraisal team shall
„ document the final findings
„ document the rating outcome(s)
„ document the Appraisal Disclosure
Statement (ADS)
„ The ADS is a summary statement describing the
appraisal
app a sa results
esu ts tthat
at includes
c udes tthe
e co
conditions
dto s a and
d
constraints under which the appraisal was performed.
„ The ADS contains information considered essential to
adequately interpret the meaning of assigned maturity
level or capability level ratings.
„ The ADS is prepared by the appraisal team leader and
provided to the appraisal sponsor.
„ A template for the ADS is provided in Appendix A.
SCAMPI - Chang-Hyun Jo @ CSU Fullerton 47

SCAMPI
Published Appraisal Results

Source:
http://sas sei cmu edu/pars/pars aspx
http://sas.sei.cmu.edu/pars/pars.aspx

24
 Source: http://sas.sei.cmu.edu/pars/pars.aspx

SCAMPI
Published Appraisal Results

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 49

Source: http://sas.sei.cmu.edu/pars/pars.aspx
SCAMPI
Published
Appraisal Results

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 50

25
References
„ Standard CMMI® Appraisal
pp Method for Process Improvement
p
(SCAMPISM) A, Version 1.2: Method Definition Document (MDD),
CMU/SEI-2006-HB-002,
http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009.
„ CMMI Appraisals @ SEI/CMU,
http://www.sei.cmu.edu/cmmi/tools/appraisals/materials.cfm, 2009.
„ CMMI @ SEI/CMU, http://www.sei.cmu.edu/cmmi/, 2009.
„ Chrissis, Mary B., Konrad, Mike, and Shrum, Sandy. CMMI:
Guidelines for Process Integration and Product Improvement, 2nd
Edition Addison-Wesley,
Edition, Addison-Wesley 2007.
2007 (ISBN 0-321-27967-0)
„ CMMI for Development (CMMI-Dev), v1.2, CMU/SEI-2006-TR-008,
ESC-TR-2006-008, SEI/CMU, August 2006.
http://www.sei.cmu.edu/library/abstracts/reports/06tr008.cfm (2009)
„ Humphrey, Watts S. Managing the Software Process, Addison Wesley,
1989. (29th Printing, 2003) (ISBN: 0-201-18095-2)

SCAMPI - Chang-Hyun Jo @ CSU Fullerton 51

26