ProblemAI
Optional AI-driven problem management for GLPI
Cluster recurring incidents, draft structured root cause analysis, and deflect known errors as tickets are typed.
What is ProblemAI?
ProblemAI automates the Problem Management practice inside GLPI. It helps teams spot recurring incidents, run structured root cause analysis, and stop the same issue being logged again.
AI drafts are labelled for transparency (EU AI Act). PII is redacted before anything is sent to a model. Humans confirm automated actions — the plugin assists; it does not silently close problems.
What It Does
Semantic clustering
Groups similar incidents so underlying problems surface instead of ticket noise
Structured RCA
5 Whys, Ishikawa (fishbone) and Pareto frameworks — not a generic summary
Incident deflection
Suggests related problems and known errors as analysts type a new ticket title
Flash alerts
Warns when similar tickets spike — useful for outages and major incidents
Privacy by design
PII redaction before model calls, audit trails, and clear “drafted by AI” labelling
ROI tracking
Cost of problem versus cost of fix, deflection metrics and token budget controls