Tech Views – Data as a Critical Service in Acquisition

We all understand the importance of data to manage any program, whether it is estimated data for planning purposes or actual data for monitoring project progress and performance.  The importance of data in managing systems acquisition programs, from a software intensive systems perspective, is particularly crucial when:

• Acquisition program requirements are complex and diverse
• Large and complex software and systems development teams are being used, including teams of teams
• Users specify requirements and accept systems, but are not part of the development process
• Complex and difficult to understand system architectures are used
• Many complex legal issues exist (e.g., contracts, copyrights, warranties)

Cost data, in particular, plays a key role in acquisition planning, making affordability trade-off decisions, and monitoring acquisition performance.  Cost estimation is a key activity underlying all of systems acquisition management, where in many cases estimates need to be made long before critical decisions have been finalized or much is known about all of the key requirements.  Historical cost data, or models built on historical data, are used to generate cost estimates.  However, interpreting what is in the data, and the relationships of data to outcomes, remains a key challenge.

This issue of the DACS Journal looks at the role of cost estimation and data in systems acquisition management.  We have assembled a group of leading subject matter experts in systems acquisition and cost estimation to share their knowledge.
Our lead article, The Need for “Acquisition Visibility”, written by Mr. Mark Krzysko, the Deputy Director of the Enterprise Information and Office of the Secretary of Defense Studies, provides an outstanding discussion and sets the stage for describing the need for data – data as a service – to support the “Better Buying Power” initiative of the DoD, and how that need is being addressed.

The DACS has, for many years, endorsed and implemented this vision of data as a service.  With this issue of the DACS Journal, in the article The DACS Software & Systems Cost and Performance Analysis Toolkit, we are announcing the addition of a new data service and resource to provide representative cost data to support the software intensive systems acquisition and cost estimation communities, in line with the vision of access to key data within the Better Buying Power initiative.  We are partnering with the University of Southern California (USC) Center for Systems and Software Engineering to develop this toolkit, leveraging their knowledge in development of the COCOMO-based models.

Several parametric cost estimation experts provide their insight on effective cost estimation methods using historical data throughout the systems acquisition lifecycle:

Throughout the systems acquisition lifecycle, the Government has to develop independent systems cost estimates.  William Roetzheim, in his article Parametric Modeling to Support Systems Acquisition, describes the use and benefits of parametric models for estimation, starting from the very earliest phases of the lifecycle.

Arlene Minkiewicz of PRICE Systems, and a frequent author for the DACS Journal, describes a strategy for being able to defend cost estimates in her article Selling Your Cost Estimate.  Having good data is an important factor.

Another article describing the importance of historical data and its use for estimation and understanding of the impacts of changes is the article by Kate Armel, from Quality Software Management, Inc., entitled History is the Key to Estimation Success.  In this article she shows, from a database of over 1,000 historical projects, the impact on cost and quality of small changes to project team size.

We also include an article addressing current research in cost estimation.  A serious challenge to estimators is developing credible estimates well before a system solution is conceived.  Addressing this, the SEI’s Cost Estimation Research Group describes their work in An Innovative Approach to Quantifying Uncertainty in Early Lifecycle Cost Estimation.  In it, they describe their proposed methods for pre-Milestone A cost estimation.

In the article Outcome-Based Software Acquisition, Roger Stewart and Lew Priven describe the use of standards-based Inspections and supporting tools to achieve an auditable and actionable outcome-based acquisition process.

We close out this issue of the DACS Journal with the article A Roadmap for Reforming the DoD’s Acquisition of Information Technology from John M. Gilligan, former Air Force Chief Information Officer (CIO), who describes an effective five-point program to improve DoD IT acquisition processes.