Partnering with one of the leading global security and aerospace companies, we built a tool to design better models where fewer errors would be uncovered during late-stage testing. The idea was to prevent engineers from going down the wrong development path, but a different discovery was made. One fateful meeting would change the trajectory of QRA Corp.
This was the response of one of the aerospace engineers when we described what our tool could do to improve modeling. We knew it could do it because it had for months with our partners on other large-scale system integrators. So, we offered to prove ourselves. If they thought our verification engine improbable, why not put it to the test. They gave us their system models and requirements, and we showed them what our tool could do. Unbeknown to us, they had intended to plant some faults in their systems design to assess the tool, but they forgot to doctor the models and so there were no faults to be found… Except, we did find faults.
Over the next hour, we had first row seats as we witnessed engineers argue about what constitutes an error and the origin of the error. We realized that to detect errors as early as possible, we should check for errors as early as possible – before any implementation was designed. We needed to develop a new tool to analyze requirements and alert authors of ambiguous, overly complex, and fundamentally malformed engineering requirements. We had our accidental discovery, and QVscribe was created. What we learned, first with these engineers, and many times since, is that at least 50% of the problems lie with the requirements, not the designs.