Glossary · Evidence, assurance and traceability
Evidence architecture
In safety assurance, an evidence architecture is the structure connecting requirements, hazards, controls, tests, documentation, operation, and changes into a coherent assurance case. It defines which item types exist, how they link and where each is stored.
- Systems engineering
- Technical documentation
In one sentence
An evidence architecture defines how requirements, hazards, controls, tests and documents are structured and linked into one assurance case.
Example
A machine builder defines an evidence architecture in which every hazard has an ID, links to one or more safety functions, each safety function links to tests and user information, and each link records its type and version.
How it applies
- Evidence: Design the evidence architecture before the project produces evidence. Retrofitting links at audit time is slow and error-prone.
- Systems engineering: It is an information model for evidence: item types, attributes, trace link types and ownership.
- Technical documentation: Include user information in the architecture. Warnings and instructions are controls whose correctness must be traceable to hazards, like guards and safety functions.
- Change control: Operation and changes are part of the architecture. A change record should link to the items it affects so that the assurance case stays current.
Evidence architecture vs. evidence package
The evidence architecture is the reusable structure. An evidence package is a concrete set of evidence assembled within that structure for one claim, function or change.