Automated Testing of an Industrial Supervisory Control and Data Acquisition (SCADA) Software System.

摘要

This MSc project involved the application of two empirical software engineering approaches, case study and action research, during the systematic testing of a part of an evolving industrial SCADA software system. We performed black-box unit testing (BBUT) on this system according to the developers' requirement where we assessed the behavior of 57 functions from the system-requirements perspective. Because of the inefficiency of available testing tools in our context, a customized BBUT tool, AutoBBUT, was developed. This was utilized as a part of this research to automatically generate NUnit test cases satisfying equivalence class partitioning, boundary value analysis and n-way testing. Afterward, we applied two white-box testing techniques, code coverage measurement and mutation testing to verify and enhance the strength of these test cases. Finally, effectiveness of symbol and branch coverage was empirically evaluated based on their fault detection ability; branch coverage was observed having more fault detection ability.