Glossary · Verification and validation
Post-deployment monitoring
Also known as: post-release monitoring, in-service monitoring
In systems engineering, post-deployment monitoring is monitoring after release or commissioning to detect safety, reliability, security, or evidence issues. It checks whether the assumptions behind the release decision still hold in operation.
- Verification
- AI
In one sentence
Post-deployment monitoring watches systems after release for safety, reliability, security and evidence issues that testing could not reveal.
Example
A manufacturer of AGVs collects anonymized stop events and near-miss reports from customer sites and reviews them monthly against the assumptions in its risk assessment.
How it applies
- Evidence: Define in advance what is monitored, how often and what triggers action. Link findings to the claims and assumptions they affect.
- Cybersecurity: Monitoring for new vulnerabilities and handling them through patch management is part of post-deployment work. Under the Cyber Resilience Act, reporting of actively exploited vulnerabilities applies from September 11, 2026, and vulnerability handling over the support period is part of the main obligations that apply from December 11, 2027.
- AI and retrieval: Systems with machine learning components can change behavior as input data shift (AI drift). The EU AI Act requires post-market monitoring for high-risk AI systems.
- Technical documentation: Monitoring results should flow back into instructions, warnings and the technical file when they reveal new hazards or misuse.
Post-deployment monitoring vs. periodic verification
Periodic verification checks that installed protective measures still work. Post-deployment monitoring looks across the fleet and over time for issues in design, assumptions, security and documentation.