Stephen R. Jernigan

Application of the SEPA methodology and tool suite to the National Cancer Institute

In health care, as with many domains, a lack of automation has hindered its ability to accomplish its primary objectives. The paper introduces a development methodology and tool suite which provide a domain based approach to system development that addresses differing client perspectives and fosters reuse. The Systems Engineering Process Activities (SEPA) methodology emphasizes early analysis and requirements gathering activities to provide a sound foundation for object derivation. Attention paid to these early analysis activities facilitates maintaining traceability and verification of deliverables. The paper explores an example application of SEPA to the National Cancer Institutes current Protocol Authoring Tool development effort.

The Systems Engineering Process Activities (SEPA) - supporting early requirements analysis and integration prior to implementation design

The SEPA methodology and its supporting tool suite address critical issues for software development practices: traceability between requirements, design and implementation; requirements reuse and code reuse; and systems integration. SEPA focuses on requirements analysis and integration prior to implementation design by supporting the capture of a spectrum of user inputs/requirements that are narrowed, refined and structured into a system design. User inputs require refinement for a number of reasons, including the need to (1) merge inputs from multiple sources, (2) discard irrelevant information, and (3) distinguish between general domain requirements and those relating to a specific implementation. Tools currently under development support (i) synthesizing requirements into a functional domain model, (ii) deriving object-oriented classes from the domain model, and (iii) producing a system design specification satisfying functional, performance and infrastructure requirements.