Developing a traceable domain reference architecture to support clinical trials at the National Cancer Institute. An experience report

Presents the requirements gathering and software architecture derivation approach developed by the University of Texas at Austin and leveraged by the National Cancer Institute (NCI) in their efforts to automate the creation, management and evaluation of clinical trials. NCI must face the complexity of managing clinical trials and of coordinating large numbers and varied types of stakeholders. SEPA (Systems Engineering Process Activities), from the University of Texas at Austin, is well-suited to address domain modeling and software development at NCI due to a strong emphasis on explicit traceability from a derived architecture to individual knowledge acquisition sessions, as well as facilitated resolution among conflicting stakeholder contributions. Specifically, this paper focuses on the SEPA Domain Reference Architecture (DRA), a software architecture designed to capture domain requirements (i.e. domain data, functionality and timing).