Software architecture based inter-feature analysis of safety and security interactions

Due to the shift towards assisted and automated driving, the stringent safety and security requirements for these systems continue to increase. Development of resulting safety and security features are often carried out independently as per their respective automotive standards. These features, when integrated into the system, can lead to unintended functional interactions between them. To improve the software quality and dependability of safety critical systems, we propose the novel method and corresponding tool I-FASST that automatically analyze UML-based software architecture models and identify inter-feature interactions amongst safety and security features. Using the I-FASST tool, we evaluate the I-FASST method by conducting a case study on a real-world driver assistance system. The evaluation shows that the I-FASST tool identifies 55 feature interactions of which 20 are of interest to safety and security experts as they exhibit potential for either synergies or interferences.