Two types of failures, two strategies
Systematic failures originate from mistakes in specification, design, implementation or tools. They cannot be handled by statistics, so they are controlled through process compliance: a rigorous, documented development lifecycle with requirements traceability, reviews, verification and validation, scaled to the required integrity level (ASIL in automotive). Random hardware failures occur unpredictably during operation, for example a bit flip or a broken connection. They cannot be prevented entirely, so they are managed with safety mechanisms such as diagnostics, redundancy, monitoring and error detection, which detect faults and bring the system to a safe state. Their effectiveness is quantified by hardware architectural metrics and failure rates.The safety case: your evidence
The safety case is a structured argument, backed by evidence, that your product is acceptably safe for its intended use. It typically contains the safety goals from the hazard analysis, the safety concept and requirements, the results of safety analyses (FMEA, FTA, FMEDA), verification and validation reports, confirmation measures and audit results, and proof of process compliance. Its value goes beyond documentation. Assessors and customers use it to decide on release for production, OEMs and Tier-1 suppliers build their own safety arguments on it, and it demonstrates your diligence if questions arise after launch.Ready to discuss your project? Tell us about your functional safety needs and we will get back to you within one business day.