How to Track OEE on the Factory Floor (Without an MES Project)

Most OEE tracking projects die the same way: a consultant quotes a six-figure MES, the plant manager flinches, and everyone goes back to spreadsheets. But tracking OEE does not require an MES. It requires three things: a way to know what state each machine is in, a way to record why it stopped, and a way to count what it produced.

This guide covers the architecture of a lightweight OEE tracking system: the one a plant can be live with in days, not quarters.

The Three Data Streams You Actually Need

1. Machine state (for Availability)

Every machine is always in one state: running, idle, breakdown, changeover, planned stop, and so on. Availability comes from how long the machine spends in each state during planned production time.

The cheapest signal is the machine's own status: a PLC output, a contactor, a run lamp, or a mold-close sensor. A digital input reading that signal gives you the state transitions with timestamps, no PLC programming required.

2. Downtime reason codes (for the "why")

State tells you the machine stopped. Reason codes tell you why, and that is the data plant managers actually act on. The practical pattern: when a machine enters a non-running state, an operator taps the reason on a small panel, a handheld, or the mobile app. Five to eight codes per line, in the operators' own language: "No material," "Setup," "Mechanical fault," "Waiting QC," "Break."

3. Production counts (for Performance and Quality)

• Performance needs unit counts or cycle time against operating time, which catches micro-stops and slow cycles that downtime logs miss entirely.
• Quality needs good vs scrap counts. A scrap counter (or the operator logging rejects on a handheld) covers it.

The Classification Layer

Raw signals become OEE data through a classification layer. The standard floor states map to the OEE components like this:

State	OEE impact
Running	Productive
Planned stop	Not counted (reduces planned time)
Changeover	Availability loss
Breakdown	Availability loss
Micro-stop	Performance loss
Slow cycle	Performance loss
Process defect	Quality loss

The subtle but critical detail: micro-stops and slow cycles are Performance losses, not Availability losses. A machine that runs 420 minutes of a 480-minute shift looks 87.5% available, but if it only produced 80% of its ideal output during that time, real OEE is far lower. Logging only downtime blinds you to the biggest hidden loss. This is exactly why the OEE formula multiplies three factors instead of just reporting uptime.

Hardware vs No-Hardware

OEE tracking splits into two deployment styles:

• Sensor-based: a small IoT node reads the machine's digital status signal and streams it wirelessly. Add an LED tower lamp that broadcasts the machine state across the floor, plus a handheld scrap counter. This is fully automatic; the operator never touches the system for Availability.
• App-only: operators log states and reasons from a mobile app. Zero hardware, works anywhere, but depends on operators tapping, and operators forget to tap.

The honest guidance: start app-only if you want zero capital cost, but budget for sensor nodes on the machines that matter most. Automatic state capture is what makes the data trustworthy enough to act on.

The Rollout That Works

1. Pick 5–10 machines, not the whole plant. One line, one shift. Prove the system before scaling.
2. Define reason codes with the operators, not for them. A 15-minute session per line. Use their words.
3. Wire the state signal. Sensor node on the PLC contact, mold-close sensor, or run lamp, whichever the machine exposes.
4. Go live for a week. Do not tune anything in week one. Just collect.
5. Review the first week's data with the plant manager. This is where the value lands: the first "we lost 40 minutes of Performance to micro-stops on line 2" moment.
6. Expand machine by machine. Add the next line once the first one is trusted.

What You Get at the End

A working OEE system gives the plant manager three things a spreadsheet never will:

• A live floor view: every machine's state, color-coded, from a phone, tablet, or browser.
• Downtime Pareto: the top reasons ranked, so improvement effort goes where the minutes are.
• Shift reports: OEE, Availability, Performance, Quality, and scrap per machine per shift, without anyone compiling them.

Voltrus OEE is built exactly this way: a sensor node reads the machine signal, the floor statuses classify automatically, reason codes and scrap are operator-friendly, and the dashboard (OEE and scrap Pareto at site, building, and machine level) appears live. Hardware is optional, setup is days, and there is no MES-shaped project anywhere in it.

Keep reading

Guide OEE Formula Explained: Availability × Performance × Quality With a Worked Example August 16, 2026 Guide How to Track OEE on the Factory Floor (Without an MES Project) August 16, 2026 Guide Manufacturing Downtime Tracking: How to Measure, Classify, and Reduce Stops August 16, 2026