Advantages of the Alternative Method for Random Hardware Failures Quantitative Evaluation - a Practical Survey for EPS 2013-01-0190

Two methods are allowed in ISO 26262-5 for hardware analysis of random hardware failures. The 1^st method is called “Evaluation of Probabilistic Metric for random Hardware Failures”. The 2^nd method is called “Evaluation of each cause of safety goal violation”. Advantages of the 2^nd method during development of ASIL D Generation 3 Electric Power Steering are presented in this paper. A reliability analysis is one of the important prerequisite for the hardware analysis and this paper shows the best practice for hardware part failure rate estimation using industry standards such as IEC TR 62380. The equally important focus is on a diagnostic coverage of each safety mechanism with respect to residual faults and with respect to relevant dual/latent point faults because any safety design can either benefit from low failure rates or from high diagnostic coverage of safety mechanism to mitigate faults. FMEA is highly recommended by ISO 26262-5 as a part of hardware analysis. There is an analogy between the 2^nd method for hardware analysis and FMEA. The 2^nd method may be considered an extension of standard FMEA techniques that also analyze the system effect of faults including SW interactions.