Modeling the Evolution of a Science Project in Software-Reliant System Acquisition Programs

摘要

Analysis work by the Software Engineering Institute (SEI) on data collected from more than 100 Independent Technical Assessments (ITAs) of software-reliant acquisition programs has produced insights into some of the most common ways that programs encounter difficulties. This paper describes work done at the SEI that is based on these insights, and intends to mitigate the effects of both misaligned acquisition program organizational incentives, and adverse software-reliant acquisition structural dynamics, by improving acquisition staff decision-making. The research presented here uses a preliminary system dynamics model to analyze a specific adverse acquisition dynamic concerning the poorly controlled evolution of small prototype efforts into full-scale systems that is called "The Evolution of a Science Project." The system dynamics model of the behavior is described, along with the results of simulations run using the model. The paper concludes with a set of lessons learned, as well as potential future research directions.