Glossary · Fault behavior and system states
Fallback
In systems engineering, a fallback is a defined alternative behavior used when the preferred function or dependency is unavailable. It is specified in advance, not improvised at runtime.
- Integrated systems
- Functional safety
In one sentence
A fallback is the planned alternative when a function or dependency fails; it must be specified, verified and made visible to operators.
Example
When the vision system of a packaging line fails, the line falls back to a slower mode with manual label checks instead of continuing to run unchecked.
How it applies
- Integrated systems: For every critical dependency, state the fallback, the condition that triggers it and how long the system may stay in it. A fallback that relies on another system creates a new dependency that has to be analyzed too.
- Functional safety: A fallback must not increase the risk. If no acceptable alternative exists, the correct fault reaction is a transition to a safe state, not a fallback.
- Verification: Fallbacks are rarely exercised in normal operation, so test them explicitly, for example with fault injection.
- Technical documentation: The operating manual must tell operators how they can recognize that the machine is running on a fallback and which limits or additional tasks apply.
Fallback vs. degraded mode
A fallback is the alternative behavior. Degraded mode is the operating condition with reduced capability that often results from it.