SEI Background:
The Carnegie Mellon Software Engineering Institute (SEI) was founded in the US in 1984 as a federally funded research and development center. The SEI works closely with defense and government organizations, industry, and academia to continually improve the development of software-intensive systems, by recognizing and promoting promising research and best practices. SEI programs focus on software engineering, computer security, and process improvement. The SEItests and refines technology through pilot programs, then disseminates proven solutions through training, licensing, and publication of best practices. The SEI’s activities fall into three areas:
Within these broad areas of work, the SEI has defined specific initiatives and projects that help organizations to acquire, build, and evolve software-intensive systems predictably on time, within expected cost, and with expected functionality.
Over the past seventeen years, DNV/Q-Labs has sponsored multiple SEI affiliates. The SEI affiliate relationship: provides DNV/Q-Labs the opportunity to influence the outcomes of various SEI initiatives, enables the SEI to have access to additional resources, and allows the affiliate to gain knowledge as well as to transition the knowledge to their home organization.
SEI Initiatives with DNV IT Global Services participants:
Capability Maturity Model Integration (CMMI) Product Suite v1.2:
The CMMI Product Suite is comprised of the CMMI Model, the Standard CMMI Appraisal Method for Process Improvement (SCAMPI) Method, and CMMI and SCAMPI Training curricula.
The purpose of the CMMI Version 1.2 piloting project was to evaluate the effectiveness and usability of the revisions to the V1.1 CMMI Product Suite prior to public release as V1.2. The results of the piloting project enabled further refinement and enhancement of the CMMI Product Suite, reduced the risk of releasing flaws, and ascertained how well the product suite met the user community’s needs.
The piloting project evaluated the changes to each individual component of the product suite. All changes to each of the components were identified, evaluated, and prioritized. The piloting project focused on the changes to those components that were associated with the greatest risks. Pilots of changes to the CMMI Product Suite components were accomplished through a variety of methods (i.e. training, appraisals, and reviews). Multiple pilots were conducted for each of the components. Pilot results were collected utilizing various methods (i.e. interviews and questionnaires). The results were summarized, and analyzed. Based on the analysis, appropriate change requests were generated to address identified defects. A draft of the CMMI model was released for pilot use. A pilot release of the SCAMPI MDD and ARC was created for appraisals of participating pilot organizations. A pilot release of the Introduction to CMMI course was created, as well. The model was piloted by using it on SCAMPI Class A appraisals and also reviewed by experts. The SCAMPI appraisal method was subject to expert review. The revised training courses were subject to expert review. Feedback was collected, analyzed, and reported at the 2006 SEPG and ESEPG Conferences. Marilyn Phillips helped organize the piloting activities.
CMMI for Acquisition (CMMI-ACQ):
Piloting of the CMMI for Acquisition constellation was also conducted in 2006. The draft model was subjected to expert review. Feedback was collected and evaluated in a similar fashion to that of the CMMI Product Suite.
CMMI for Services (CMMI-SVC):
Frank Niessink of DNV/CIBIT participates on the CMMI for services team, multiple ITGS consultants participate as members of the review team (Annie Combelles, Sophie Balous, Siegfried Raschke, Marilyn Phillips).
Lead Appraiser Body of Knowledge (LA-BOK):
Anne-Sophie Luce and David Card DNV/Q-Labs are participating in working groupd to define a body of knowledge for SCAMPI Lead Appraisers. The body of knowledge will be used to improve the criteria and testing required for certification of Lead Appraisers. This has proven to be especially important for high maturity appraisals, where additional quantitative and statistical knowledge is important.