Systems engineering for software-intensive projects

Abstract

The technical shortcomings of major software-intensive projects, as well as their cost overruns and schedule slippages, are usually attributed to poor management practices. We hope to show that the application of established systems engineering techniques can help to overcome, or at least to reduce, the difficulties that are encountered in the development of large information processing and automated control systems. The paper focuses on two life cycle events that have been identified in government studies as causes of the unsatisfactory outcomes: requirements formulation and early assessment of system capabilities. It is shown that establishing a figure of merit (FOM) for the primary system function at the outset of development helps in the management of requirements as well as in the early assessment of the system capabilities. As an example of trade-off methodology, an important system engineering tool that can be applied throughout the life cycle is also presented.