Doctoral student at MDH improves possibilities to test safety-critical systems

Nov 19, 2014 | Research/Cooperation

Eduard Enoiu is carrying out doctoral studies on safety-critical systems at Mälardalen University (MDH) and has developed a tool that renders the testing of such systems more effective. On Friday, 7 November, he defended his licentiate thesis, where he presented the method behind the tool.

A safety-critical system is a system that can cause a disaster if it stops working. An example is the control system of a train, which Eduard Enoiu chose to focus on in his research. This kind of system is becoming more and more complex, and today it is tested by someone who manually creates a test case for it – a method that is not only time-consuming and expensive, but also susceptible to human error.

– I wanted to find a better way of testing systems. If all the weaknesses are to be detected, tests need to be carried out earlier and more frequently in the development process, and with automatized tools, says Eduard Enoiu.

CompleteTest

The outcome was CompleteTest, a tool that generates test cases automatically and in just a few seconds.

– Even if CompleteTest uses a so-called model checker, which is a complex technology requiring a high degree of expertise, our tool offers a simple interface for the intended users. The process becomes less tricky and faster to learn, which will make things more efficient. If it works in its intended environment in the way we believe and hope it will, it can be used to detect software errors in a train’s control system, says Eduard Enoiu.

While developing the tool, he carried out case studies together with Bombardier Transportation AB, where the method has been applied to real programs. The results suggest that it is time-effective and well-functioning, but that the technology is not ready yet to be applied to more complex programs. In particular, Eduard Enoiu thinks that he needs to learn more about what function CompleteTest fulfils in its specific area of application and for the people working with it.

– One of the things I have learned while working with my licentiate thesis is that research on software testing is still in its infancy when it comes to our understanding of how to use it in practice. By paying attention to this aspect in our future research, we can start to ask questions we hadn’t thought of previously. Ultimately, it is all about using our knowledge to make software testing less complex.

Appreciates the “double life”

Eduard Enoiu plans on continuing his research regarding embedded systems, on developing his tool and evaluating how it performs in practice. This, he hopes, will result in a doctoral thesis in 2016.

– A great thing about research is that you rarely know exactly what you are going to work with next week. At the beginning of my career, when I was younger, I thought that was a little scary. But during my studies at MDH, I have realized that the most rewarding aspect of research is the clash with the new, the unexpected – that is what keeps me intellectually stimulated. So I want to continue living a double life of computer engineer and scientist. Research adds a dimension to my professional life that I think I’ll appreciate more and more the older I get.

He thinks that the future of software development looks promising.

– If you are well educated and use your competence well, you can change the world before you’re 35.