Reusing safety-critical systems saves time and money
Jul 13, 2015 | Research/Cooperation
MDH researcher Irfan Sljivo is a doctoral student in Embedded Systems at Mälardalen University (MDH) and has developed methods which facilitate the reusing of software in safety-critical systems. This leads to reduced costs and a more structured working method for manufacturers.
Cars, trains and aeroplanes are examples of safety-critical systems. These can cause great damage if they do not work as they should. Researcher Irfan Sljivo focuses on safety-critical systems in transport vehicles, where software faults can have disastrous consequences, and at worst lead to accidents. These systems are becoming increasingly more common and increasingly more complex, which renders it more difficult to guarantee their safety.
– It often costs more to ensure that the safety-critical systems in transport vehicles work than what it costs to develop them, he says.
"All systems can be unsafe in the wrong conditions"
Irfan Sljivo wants to make it possible to reuse software components of safety-critical systems, which is troublesome in the current situation. Every branch has a specific safety standard that the systems have to follow, which means that software from one type of vehicle cannot easily be reused in another. Since the software here is working in new conditions there is no longer any guarantee that the system is safe.
– I want to facilitate the reuse of software to reduce the work time and the cost of adapting it according to the prevailing safety standards. By mapping what is called safety contracts, which guarantee that a certain software component follows the safety requirements in a certain situation, this can be reused between different models without any risk of system errors, says Irfan Sljivo.
Results with wide application
In his Licentiate thesis he presents how these safety contracts are developed and used. The methods have been tested by case studies, in which they have been demonstrated in safety-critical systems in wheel loaders. The results are relevant for the reuse of software components in all vehicle branches, and one future research goal is to make the reuse of software possible between different types of vehicles, for example from cars to aeroplanes.
– Reduced costs of adapting systems according to safety standards implies increased possibilities of more detailed safety analyses, and in turn to greater confidence in the safety of the systems. Moreover, resources can be directed to the development of system functions, which has previously not been financially possible, says Irfan Sljivo.
Irfan Sljivo has a Master’s degree in Electronics from the University of Sarajevo. It was during an exchange semester that he first came into contact with Sweden and MDH.
– I enjoyed the working climate, the structure of the course and the close connection with industry. That is the main reason why I returned to MDH to start research as a doctoral student, he says.
The next step is to continue his research towards a doctoral thesis, now with the focus on safety-critical systems in autonomous vehicles such as cars, trucks and industrial machines.
– These rely even more on software than what vehicles do today. The interaction between the artificial intelligence and the security in these systems feels very exciting, says Irfan Sljivo.