Platform · Operate
Bring the plant floor into the governed platform, safely.
Live machine signals over OPC UA and MQTT Sparkplug B, normalized to one asset model, with a safety boundary that is never addressable.
What it is
The factory floor has been walled off from the back office for decades. Machine telemetry brings it in: live signals from your machines, normalized to an ISA-95 asset model, visualized as gauges and history, and available to the same agents that reason over your business systems. The safety boundary is load-bearing: the Safety Instrumented System is never addressable, and any write goes through MES or SCADA with approval.
Sound familiar?
The work nobody signed up for.
If you have said one of these this month, the rest of this page is for you.
The PLC knows the spindle load. The planner finds out from the operator at break.
Our historian keeps everything and shows nothing.
I asked for vibration spectra and got a CSV of hourly averages.
Connecting one more machine means a vendor visit and a quote.
Every machine has its own screen. There is no screen for the line.
Use cases
What you would use it for, first.
Each one starts from a problem you can name, and ends with something a person can see has changed. Live means it runs in the product today; pilot means we build it with you.
- 01
Live gauges for every machine
Live todayTodayMachine state is read off a local panel, one machine at a time.
With the platformSignals arrive live over OPC UA or MQTT Sparkplug B through the edge and appear as gauges per machine: temperature, vibration, speed, flow, level and throughput.
What changesThe shift supervisor sees the line on one screen.
- 02
Spectra, not averages
Live todayTodayAveraged vibration hides the frequency where a bearing defect shows first.
With the platformEach machine keeps its vibration spectrum with the dominant frequency marked, next to 24-hour to 90-day history for every signal.
What changesAnalysts see the defect frequency climbing, weeks before the average moves.
- 03
Every signal in one asset model
Live todayTodayThe same temperature is called three different things on three machines.
With the platformSignals land in an ISA-95 asset model with quality and timestamps, so agents and people read them the same way across the plant.
What changesA rule written for one oven works on the next one.
- 04
Modbus, MTConnect and controllers from the catalog
Built in a pilotTodayOlder machines and CNCs do not speak OPC UA or Sparkplug B.
With the platformThese protocols are in the catalog on sample data today. In a pilot we build the edge connection for your machines.
What changesThe older half of the plant joins the same model.
Business value
Where the return comes from.
The levers, and the range other organizations have published for each. Ranges are typical of the category, not a promise and not our customers' results.
Downtime hours
−35–45%The upper-end range from DOE-compiled surveys (2010). Pair it with Deloitte's more conservative figures.
US Department of Energy (FEMP), Operations & Maintenance Best Practices Guide, Release 3.0 (2010)Cost of an hour, rising
4x since 2019 in heavy industryVendor study. Each hour of warning is worth more every year.
Siemens, The True Cost of Downtime 2024 (2024)Industry-wide downtime cost
$50B a yearDeloitte's 2017 estimate for industrial manufacturers, citing a secondary study.
Deloitte Insights, Predictive maintenance and the smart factory (2017)Data-team time spent on extracts
History and spectra export in one click, shaped for models. Measure it on your next request.
Run your numbers
Unplanned downtime you could avoid.
Every input is yours to change, and the arithmetic is shown in full. Nothing is hidden in a multiplier.
324 × €25,000 × 25% = €2,025,000
An estimate on the numbers above, not a measurement and not a customer result. In an Operations X-Ray we replace the assumptions with figures measured on your own systems.
Before and after
The same week, run differently.
Machine state is read off a local panel.
Every machine has live gauges in one place.
Vibration arrives as hourly averages.
Vibration arrives as a spectrum with the dominant frequency.
Each machine names its signals differently.
Signals share one asset model.
A data extract is a ticket and a week.
Ninety days of history is a CSV export.
Capabilities
What you get.
Live today over OPC UA and MQTT Sparkplug B, through edge adapters that speak the real wire protocols. Modbus, MTConnect and vendor controllers are in the catalog on sample data and are built for your stack in a pilot.
Signals normalize to an ISA-95 model, so a temperature or a vibration means the same thing across machines.
Per-machine dashboards with live gauges, long-range history and vibration spectra, plus a training-data export.
The Safety Instrumented System is never addressable. Any change is routed through MES or SCADA with approval, never a direct PLC write.
How it works
Three moves.
An edge adapter subscribes to machine signals over standard protocols.
Signals land in one ISA-95 asset model with quality and timestamps.
Agents use the telemetry to diagnose and, within governance, propose a correction through MES.
Why ours is different
The part competitors leave out.
- OT and IT share one platform and one governance model, not two silos.
- The safety block is structural, so an unsafe plant-floor action is not expressible.
- Telemetry feeds the same agents and graph as your business systems, closing the OT to IT loop.
Questions
Frequently asked.
Only through MES or SCADA, and only with approval at the right risk tier. A direct PLC or Safety Instrumented System write is not expressible in the architecture.
Live today: anything reachable over OPC UA or MQTT Sparkplug B. Modbus, MTConnect and vendor controllers are catalog entries on sample data, built for your stack in a pilot.
More platform capabilities
Operations X-Ray
Connect one machine of yours.
If it speaks OPC UA or MQTT Sparkplug B, we bring it in live and show its gauges, history and spectrum.