Operations Center
Live operational KPIs, alarm feed, and telemetry timeline.
The Operations Center (/operations) is the default landing view after sign-in. It summarizes plant health, rule engine activity, and live telemetry in one screen.
What you'll do
- Scan KPI cards for connectivity, alarms, throughput, and coverage
- Review the recent alarms feed and jump to full alarm management
- Watch the live telemetry timeline for incoming tag values
Requires ops.read permission (included for ADMIN, OPERATOR, VIEWER, and ENGINEER default roles).
KPI overview
The dashboard uses shadcn metric cards with muted icons, a Live badge, and Updated … ago at the top. Values stay neutral unless something is off (open alarms, low connectivity). Status uses a left-aligned progress bar and number color; ingest uses a blue sparkline (last-point marker, % change vs the series start) beside the value; alarms use a stacked severity bar.
There is no header View alarms button — open alarms still show a red banner, and each alarm card links through when you have alarms.read.
Metric cards
| KPI | Meaning | Mini chart |
|---|---|---|
| Open alarms | Active alarms needing attention | Stacked severity bar |
| Online rate | Share of active devices online in the last ~15 seconds | Progress bar |
| Ingest rate | Tag values per second (10s average); sparkline is the last ~2 minutes from the API | Area sparkline + % vs window start |
| Gateways | Edge gateways online vs total (revoked excluded) | Progress bar |
| Devices | Online vs active devices; subtitle includes plant count | Progress bar |
| Active rules | Enabled rules in the rule engine | Count only (no fake trend) |
| Programs | Desired RUNNING vs currently publishing telemetry | Count only (no fake trend) |
| Wall screens | Paired kiosk screens online vs total | Progress bar |
The page refreshes automatically every few seconds.
Platform health
Platform health shows five stats: API memory, cache memory, DB size, live tags, and pending saves. Usage bars stay neutral below 75%, amber above 75%, red above 90%.
With platform.read, View details opens Platform health. Without it, KPIs still load; only the link is hidden.
Floor snapshot and maintenance
If you have production.read, a floor snapshot card shows station count, lowest OEE, and stations in DOWN/ERROR (from GET /production/stations).
If you have maintenance.read, a maintenance card shows open work orders (OPEN + IN_PROGRESS) from the operations summary.
Recent alarms vs Alarms page
| Surface | Purpose |
|---|---|
| Operations Center feed | Last few alarms for situational awareness |
Alarms (/alarms) | Full list, filters, acknowledge workflow, history |
All alarms (outline button) appears when you have alarms.read. Use Operations Center for at-a-glance monitoring; use Alarms for triage.
Live telemetry timeline
The right panel shows the most recent telemetry events (tag path, value, quality, time ago) from the Redis live cache for active tags — not a Postgres history replay.
UNS Explorer (outline button) appears when you have tags.read. For deep history or search, use UNS Explorer.