Risk Assessment and Development Cost Optimization in Software Defined Vehicles

Abstract

Vehicle design has entered a new stage, namely, Software Defined Vehicles (SDV), where functional safety is required to be guaranteed for risk control, and development cost needs to be optimized for profit maximization. This paper targets to optimize the development cost under the functional safety requirement for a safety-aware SDV, based on the automotive safety integrity level (ASIL) decomposition defined in ISO 26262. For this, a two-stage solution is proposed, which includes functional safety risk assessment and development cost optimization. The first stage develops a new fast risk assessment (FRA) algorithm to assess the functional safety risk, including the joint reliability risk and the real-time risk, of the SDV functionality. The second stage proposes a dual requirement guarantee (DRG) algorithm to optimize the development cost considering reliability and real-time requirements jointly. Our experiments demonstrate that the proposed two-stage solution guarantees the functional safety requirement while reducing the development cost by 20%-24%.