Now live BiseraDB — Onshore's own enterprise data engine is here. See what it can do →
Robotics & Manufacturing

A machine you can’t measure is a machine you can’t improve.

Most plants estimate utilization from memory and a clipboard. The equipment that would tell them the truth is either too old to talk or locked behind a controller integration nobody has time for. Onshore Systems takes a different route: clamp a wireless sensor on the machine, read its motion, and turn any asset — press, lathe, or robot arm — into a live utilization signal in an afternoon.

Book a 15-minute demo Read the methodology
Floor Monitor  ·  site: north-plant live · 7-day window
AssetStateUtilizationCycles 24hDown 7d
Haas VF-2 machine · ruuvi-3c9e running 4182h 10m
Cincinnati brake press machine · ruuvi-a1f0 idle 969h 44m
Pick-arm 03 robot · dobot-tcp running 1,20438m
Doosan 2600 lathe machine · ruuvi-77b2 off 031h 02m
Example data — not a live customer floor. Real deployments read these numbers off the sensors.

Sense the machine from the outside

A running machine vibrates and moves. That signal is enough to know whether it is cutting, sitting, or switched off — without a single wire into the control cabinet.

1

Clamp on the sensor

A coin-cell wireless tag, magnet-mounted to the spindle, motor housing, or frame. About ten minutes per machine, no electrician.

Reads: 3-axis acceleration, an onboard motion counter, temperature, battery. Range: Bluetooth LE to a fixed gateway — one gateway covers a bay.
2

Turn motion into state

A gateway watches a short rolling window and calls each asset running, idle, or off — with hysteresis so a pause between cycles doesn’t flap the state.

Running: vibration RMS over threshold, or motion-counter rate over threshold. Off: no signal from the tag for two minutes.
3

See the floor

State intervals aggregate to hourly utilization, an OEE figure where a cycle target exists, and a downtime Pareto by reason — on any browser.

Utilization: running time ÷ observed time. OEE: availability × performance × quality, per asset and shift.

Sensor to dashboard, on infrastructure we run ourselves

Beacon Onshore is the product that carries the approach. The edge is a small gateway; the backend is the same production stack that runs Onshore’s own platforms.

Clamp-on tag BLE advertisement Gateway classify · roll up Ingest API HTTPS · keyed BiseraDB per-site namespace Dashboard state · OEE · alerts
• No inbound connection to the plant • One site’s data never shares a collection with another’s • Works behind a firewall or on a cellular gateway

The operator confirms, the AI proposes

When a model suggests a downtime reason, it lands in a review queue — a person confirms, edits, or rejects it before it counts. Nothing is auto-applied. The floor stays in control of its own record.

Robots are read-only assets

Arms and AMRs are monitored the same way as machines — state, cycles, faults, e-stop — over the controller’s own status port. Motion, trajectory, and safety stay entirely with the certified controller. Beacon never sends a command.

New equipment is data, not a migration

The floor model grows as the floor does. Add a machine type, a sensor kind, or a relationship and it’s a new record — no schema change, no downtime, no integration ticket.

One view across brand and age

A thirty-year-old press and a new six-axis robot appear on the same screen, measured the same way. No per-vendor connector, no MES in the middle.

Running on real hardware, not a slide

The reference deployment is a wireless tag on a rack in Onshore’s own server room, streaming through the same gateway and ingest path a customer would use.

~10 min
to mount a sensor and see it on the dashboard
60 s
from sensor reading to the cloud, on a one-minute flush
~13 ms
to answer a per-asset trend query at scale

The methodology, in full

How a low-cost motion sensor becomes a defensible utilization and OEE number — the signal, the state machine, the math, the storage model, and where the method stops.

Clamp-On OEE: Measuring Machine Utilization Without Machine Integration

Covers the sensing approach (accelerometer RMS and the onboard motion counter), the run / idle / off state machine and its hysteresis, the path from state intervals to availability × performance × quality, the time-bucketed storage model that keeps every query a single indexed lookup, a real deployment, and an honest account of the method’s limits.

Onshore Systems · Manufacturing & Robotics
Read the whitepaper →

Who it’s for

Logistics & 3PL

Warehouses running robot arms and AMRs that need a vendor-neutral view of fleet state and uptime.

Job shops & small manufacturers

Mixed, older equipment and no appetite for a plant-wide IoT project.

Ops leaders

Anyone who can’t get a straight answer on true machine utilization — or whose last MES/IoT effort stalled on integration.

Put it on three machines for two weeks

Book a short call and we’ll walk the live dashboard against your machine mix. If it’s a fit, we ship sensors for a free two-week pilot on three assets — you keep the data either way.

Book a demo Beacon Onshore product page →

Tier-1 monitoring only. Beacon Onshore observes equipment state; it does not command, move, or override any machine or robot controller. Dashboard figures shown on this page are illustrative examples, not customer data.