Plant Ops Copilot
loading…

Multi-agent triage for a simulated bottling line

Four machines, six hours of 10-second sensor data, at most one injected fault. Three AI agents detect the problem, look up the likely root cause in maintenance manuals and write the shift handover report. A human approves the work order. All data is synthetic.

  1. Pick a shift
  2. Explore the data (computed live, no AI involved)
  3. Run the agents and watch them work

1 · Choose a shift

Reveal what was injected (spoiler)

2 · Line data live computation · no LLM

3 · Agent investigation

How it works

Line simulator 4 machines · faults · OPC UA dbt models state runs · KPI marts DuckDB read-only Maintenance manuals BM25 search Google ADK · SequentialAgent · state passed between agents Detection agent scan_line · get_signal_trend Diagnosis agent search_manuals · get_machine_kpis Report agent report · propose_work_order SQL · stats · IsolationForest LLM Gemini or Azure Foundry FastAPI + server-sent events Azure Container Apps · scales to zero Browser live trace · human approval