Summary


Which One Should I Use?

It depends

The staged representation is like most of the currently accepted models, such as the CMM for Software. It is very useful for assessments; most recent procurements have mandated contract awards based on Maturity Level ratings. Following the staged representation produces a Maturity Level rating.

The continuous representation was designed to tailor the model and process improvement approaches, focusing on specific categories that might match those areas in your organizations where problems exist. For example, suppose your organization does only Independent Verification and Validation. You do not write software. You produce no product. You simply review, as an independent third party, those products produced by some other party and you write your report. This type of organization might want to focus on the Verification, Validation, and Product Integration process areas. The continuous representation allows this approach to take place.

One approach to take in selecting which representation to use is to consider the following two questions.

How Much Experience Does Your Organization Have with Process Improvement?

Why? Because we believe that the continuous representation requires a higher level of process improvement understanding than the staged representation. The staged representation embeds Process Area decisions that an inexperienced organization might not be prepared to make. For example, an organization just beginning process improvement usually has little idea as to where to start. The staged representation tells you to begin at the beginning ” Maturity Level 2 ” and after attaining that level, continue to the next ” Maturity Level 3. And so on. Within those levels are the process areas to include in your process improvement strategy. So, when just beginning, an organization might state that "We will begin with Maturity Level 2, which means satisfying the goals for all of the process areas in Maturity Level 2." The organization would then prioritize which process areas to focus their efforts on first (see subsequent chapters of this book for how to set up and structure your process improvement initiative). Our experience has also shown that organizations inexperienced in process improvement often put too much focus on technological solutions and not enough focus on management issues. Therefore, an inexperienced organization using the continuous representation might focus its efforts on the Engineering Processes, as opposed to those emphasizing project and process management. While improving engineering is, of course, important, without the foundation of strong project and process management, it is likely that engineering improvements would not be sustained, if accomplished at all. Those organizations that are more experienced with process improvement, and more experienced with identifying and mapping their organizational weaknesses to areas where the CMMI can offer help, would probably be able to use the continuous representation more successfully. Remember: the continuous representation allows an organization to select (that means "pick and choose") which process areas to focus on by determining which of the process areas match the organization's business needs and objectives.

What Model(s) Is Your Current Process Program Based On?

Our experience shows that an organization should build on the positive experiences it has had with other models used in that organization. The CMMI requires enough changes without introducing unnecessary confusion. So, if you have been happy with the staged structure, continue to use it. If you have been happy with the continuous structure, continue to use it. If you want to try something new, then be our guest. Just remember that whenever you introduce something new, you introduce new problems and new headaches ; but you also may stimulate and revive your organization's process improvement effort.

Exhibit 1 shows the impact of process improvement experience and current model experience on the representation decision. Four approaches are shown, as represented by the leftmost column. The four other columns represent model experience and process improvement experience. An entry with a plus (+) sign indicates a positive impact and a minus (-) sign indicates a negative impact on the approach.

Exhibit 1: Representation Selection Matrix
start example
 

Current Process Improvement Program

Approaches

Based on CMM for Software

Based on SECM or Related Model

High Process Improvement Experience

Low Process Improvement Experience

Staged Representation

+    

Builds on current experience

-    

May overlook interpretation issues

-    

Learning curve for new paradigm

+    

Defined sequence of improvements

 

+    

Easier to understand

+    

Helps maintain focus

Continuous Representation

-    

Learning curve for new paradigm

+    

Allows focus on significant area

+    

Builds on current experience

+    

Can continue with current plans

+    

Supports informed decisions

-    

Bad focus decisions likely

Start with Continuous, move to Staged

-    

Major short- term impact

+    

Allows focus on significant area

-    

Minor short-term impact

+    

Longer-term easier comparison

+    

Able to make informed focus decision

+    

Harder to build experience

Start with Staged, move to Continuous

-    

Minor short-term impact

+    

Longer-term focus on significant areas

-    

Major short-term impact

 

+    

Easier to build experience

end example
 

The two representations are not mutually exclusive. Each representation contains the same basic information ” it is just structured differently. Each representation promotes process improvement and can be used in assessments. The continuous representation allows the organization to choose which process areas to focus on, as well as which discipline to focus on (e.g., software engineering, systems engineering, IPPD, supplier management). The CMMI also suggests that tailoring of the model can be done, but

  • Caveat: The more tailoring of the model done, the less likely an organization is to achieve its improvement goals. The CMMI contains best practices of organizations. If you deviate from a best practice, you must strongly justify why and substitute another practice that satisfies the intent of the CMMI practice originally listed. Intent is like beauty ” different in the eye of every beholder. It is also difficult to define intent. Most organizations try to justify their not doing something in the model. What they really would like to write, but cannot for obvious reasons, is "It was too hard for us to do. We didn't have the money. We didn't have the people. We didn't have the time before the assessment team came in." These are not good justifications for not doing something described in the model.

There is also the concept of Equivalent Staging. What is Equivalent Staging? It is an attempt to match Maturity Levels in the staged representation to Capability Levels in the continuous representation. The levels compared are called "Target Staging." Target Staging is a sequence of targeted process areas that describe the path of process improvement the organization will take. Care should be taken to ensure that dependencies between the generic practices and process areas are implemented. This is where the organization documents the process areas (PAs) it will focus on, justifies this approach, and tracks the PAs back to business objectives (see Exhibit 2).

Exhibit 2: Equivalent Staging
start example
click to expand
end example
 

The reader should be advised that to be Target Profile 4 or 5, you do not need to achieve the Generic Goals for Capability Levels 4 and 5. That means that you need to satisfy the specific goals and practices for the process areas (Organization Process Performance ” OPP ” and Quantitative Project Management ” QPM ” for Target Profile 4, and Organizational Innovation and Deployment ” OID ” and Causal Analysis and Resolution ” CAR ” for Target Profile 5); plus, you need to satisfy the generic goals and practices for Capability Levels 1, 2, and 3 ” not 4 or 5. Careful reading of the table should lead the reader to conclude that to achieve Target Profile 3, an organization needs to satisfy the goals for all of the process areas in Maturity Levels 2 and 3 ” even if using the continuous representation. Remember that not all of the process areas contained within Maturity Level 3 in the staged representation are also contained within the process categories for Capability Level 3 in the continuous representation. Proceed along this path with caution.

As of this writing, the authors of this book know of no procurement agency that has allowed the substitution of a Capability Level for a Maturity Level for contract award.

The interesting point to make here is that the staged representation is really continuous; and the continuous representation is really staged. That is, because of all the interdependencies among the process areas and relationships that must be met among the generic goals and practices, everything is related to everything else. For example, in the staged representation, you cannot really plan your project in Project Planning (Level 2) if you have not already begun to manage your requirements in Requirements Management (Level 2). So Requirements Management feeds into Project Planning. How can you manage requirements (Requirements Management Level 2) if you have not defined them in some way, shape, or form yet (Requirements Development Level 3)? How can you effectively start a process improvement program if you have not set up a group of people and follow the guidelines in Organizational Process Focus (Maturity Level 3)? Yet that is Level 3 and the organization is just starting its process improvement program, and so is at Level 1. How can you have Verification, Validation, and Product Integration as separate entities? Generally, in most organizations, if you do one, you must do the others. The CMMI states that using the continuous representation, and focusing on Capability Levels, allow an organization to select process areas within process categories on which to focus their process improvement efforts. In the continuous representation, how can you attempt Validation, or Verification, or practically any other process area without also tackling (to some extent) Project Planning, Project Monitoring and Control, and Product and Process Quality Assurance? So, those of you thinking of selecting the continuous representation so that you can select only a few process areas for improvement will find that you are sadly mistaken. Everything is related. The different representations used are simply to help the reader understand the basic tenets of the model.




Interpreting the CMMI(c) A Process Improvement Approach
Interpreting the CMMI (R): A Process Improvement Approach, Second Edition
ISBN: 142006052X
EAN: 2147483647
Year: 2005
Pages: 205

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net