An Automated Fault Tree Generation Approach With Fault Configuration Based on Model Checking

Fault tree generation technology is a key issue for safety analysis of large complex systems. Traditional safety analysis methods usually describe the origin, propagation, or concrete behavior of the fault and do not portray the constraints between faults. However, these constraints are the system’s characteristics, and a lack of expression of these constraints will make the fault model defective, thereby resulting in a fault tree that will reduce the accuracy of the safety analysis. To improve the efficiency and accuracy of safety analysis, this paper proposes a fault tree generation method that is based on fault configuration and introduces the variability management of software product lines to model system faults and perform the formal analysis. First, the fault feature diagram is defined to describe the constraint relationships between system faults, and the fault-labeled transition system is defined based on the Kripke structure to describe the system behavior. Then, based on the model semantics, the procedure for generating fault trees by model checking is established. Finally, using temporal logic to describe the system safety attributes, we adopt the model checking tool SNIP to verify the safety attributes and generate the fault tree automatically. The fault modeling method that is proposed in this paper includes the inherent constraints between faults, which makes the system fault model more realistic and accurate. A case study demonstrates the effectiveness of the proposed method.