Glossary Updates12 new terms added to the glossaries · October 2, 2026, 22:44 CEST
AI TechDocKnowledge

Glossary · Evidence, assurance and traceability

Evidence chain

In safety assurance, an evidence chain is the linked sequence from a requirement or hazard through the design decision, implementation, verification, and user information. A break anywhere in the chain leaves the claim unsupported.

  • Systems engineering
  • Technical documentation

In one sentence

An evidence chain links a hazard or requirement to design, implementation, verification and user information without gaps.

Example

The evidence chain for a crushing hazard on a press runs from the risk assessment entry to the decision for a light curtain, its installation record, the stopping time test and the warning in the operating manual.

How it applies

  • Evidence: Build the chain with explicit trace links, not by naming conventions. Check it for gaps before release and after changes.
  • Machine safety: Start from the hazard. Every hazard that needs risk reduction should end in verified measures and, where residual risk remains, user information.
  • Technical documentation: The last link is often the manual. Structured safety messages, such as a DITA hazard statement with an ID, make that link precise.
  • Audit: Auditors typically pick a hazard and follow the chain. A chain that can be followed in minutes shows control.

Evidence chain vs. traceability

Traceability is the general ability to follow links. An evidence chain is one specific path through those links that supports a claim.

By knowledge.aitechdoc.world · Published September 25, 2026 · Last reviewed

Source: AI TechDoc Blog editorial definition, based on ISO 12100:2010, Safety of machinery — General principles for design — Risk assessment and risk reduction

Definitions follow the cited standards and specifications. Where a source is a copyrighted publication, such as an ISO, IEC or EN standard, the definition is a close paraphrase, not a verbatim quotation, so as not to infringe copyright. We recommend reading the original publication. The sections “How it applies” are editorial commentary by AI TechDoc Blog and are not part of any standard.

Seen a mistake? Send us a note!