Systems engineering is all about a structured and auditable approach to identifying requirements, managing interfaces, and controlling risks throughout the project lifecycle. The systems engineer must analyze, specify, design, and verify the system to ensure that its functionality is as per the requirements.
As markets are becoming more demanding, products are becoming more customized, and design life-cycles are getting shorter, systems engineering can provide an organization with a clear competitive advantage. The correct application of systems principles and practices will realize substantial benefits including reduced design lead time, design changes, and errors in production and improved reliability.
Despite well-defined systems engineering process, safety critical systems have been known to have issues. There have been two major issues reported in 2015. This means, we have not yet mastered the art of system and software development. We have been spending a lot of time on the right side of the V model, testing the system. Testing is approximately 80% of the activity in the system development lifecycle today. We have done all sorts of testing at various levels, yet we find errors in final stages of product testing. 70% to 80% of the errors found during testing are in the requirements. This clearly shows that there is something wrong with the way we do things today. We require looking at the V model and bringing in a paradigm shift here.
We, therefore need to shift our focus from testing on the right to the place where the most errors are found – the left side. This talk elaborates on the concept of “mathematizing the left” and “automating the right” of the V model of systems development. Model-based development, property based requirements, and formally proving the correctness is the way forward. Automating testing with coverage metrics at various levels and ensuring 100% property coverage is the way forward for the right side.
