Risk, Safety, and Human Factors
Risk, Safety, and Human Factors
This domain tests risk definition and treatment, hazard analysis methods, safety integrity concepts, and human error in reliability.
Core rules
- Risk combines likelihood and consequence. A hazard is a source of harm; risk is the chance and severity of harm.
- ALARP means reduce risk as low as reasonably practicable. LOPA evaluates independent protection layers; HAZOP applies guide words to process nodes.
- SIL and ASIL are integrity targets for safety functions, not labels for overall product reliability.
- Human error probability (HEP) is modified by performance shaping factors such as time, workload, procedures, training, and feedback. Independent checks multiply HEPs.
Worked example
A task has HEP=0.02. An independent verification step has HEP=0.10 and catches 90% of errors. Residual error≈0.02×0.10=0.002; the check reduces error by a factor of 10, not to zero. If the check is not independent because the same person uses the same cue, the multiplication is invalid and actual risk is higher.
Traps
- Treating human and hardware failures as independent when they share a cue.
- Confusing SIL with MTBF.
- Using a risk matrix without defined consequence categories.
- Ignoring latent errors revealed only during emergencies.