Systems Engineering of Computer-Based Systems, State of Practice Working Group

This working group report defines the need for a discipline devoted to engineering of computer-based systems, identifies current practice and needed research, and suggests improvements that are achievable today. uring the past decade, advances in microprocessor and network technology have led to the proliferation of complex systems with distributed ,: processing and databases, internal communication systems, and heterogeneous components. The processing components can by themselves comprise a system, or they can be embedded in a physical system such as an automobile, aircraft, or medical diagnostic system. Both the encompassing system and the processing system are known as computer-based systems (CBSs). Developing large computer-based systems with complex dynamics and component interdependencies requires analysis of critical end-to-end processing flows to determine feasibility and proper allocation. Currently, no engineering discipline provides the knowledge base for the necessary trade-off studies concerning software, hardware, and communication components; a new discipline is needed at the systems engineering level.’ Industry has recognized the size and scope of problems with systems engineering processes, and several organizations have been formed to advance the systems engineering discipline. These include the American Institute of Aeronautics and Astronautics’ Systems Engineering Working Group, the National Council on Systems Engineering, and Europe’s Atmosphere Project.? However, recognition of the need for a special discipline addressing the system engineering of computerbased systems (ECBS) is just emerging, as evidenced by the recently formed IEEE Computer Society Task Force on ECBS and recently published textbooks3 The IEEE Computer Society created its task force to promote the ECBS discipline, encourage research in the field, and establish a framework for education and training in 1991.1.s This working group report, in presenting the case for an ECBS discipline, discusses current practices and identifies market and social imperatives for their improvement by