Asil decomposition using SMT

Abstract

The ISO 26262 defines discrete Automotive Safety Integrity Levels (ASILs) to enforce functional safety. Each component in the automotive system under development must have an associated ASIL. Higher ASIL implies more development cost and effort. ASIL decomposition allows reducing ASIL allocated to components whose joint failure is the only cause for the violation of a safety goal. Fault trees are widely used in the safety analysis process and hence in the ASIL allocation. In this paper, we present a new approach for solving the ASIL decomposition problem using Satisfiability Modulo Theories (SMT). The fault tree structure is fully represented in SMT. Compared to other approaches for ASIL decomposition; our approach eliminates the need of finding the Minimal Cut Set (MCS) of the fault tree. Moreover, it does not require assigning a numerical cost value for each ASIL. Recent emerging trend in powerful SMT solvers for solving objective functions is utilized to find the optimal ASIL decomposition.