The Six Big Losses framework, developed by Seiichi Nakajima as part of Total Productive Maintenance (TPM), identifies every type of production loss that reduces OEE. The power of the framework is that it doesn't just say "OEE is low" — it tells you which category of loss is responsible, so you can apply the right countermeasure.
The Six Big Losses divide into three pairs, each corresponding to one OEE component:
What it is: Unplanned machine stops due to breakdowns — motor failure, belt snapping, sensor malfunction, electrical faults, jams requiring maintenance intervention.
Manufacturing example: A conveyor drive motor fails mid-shift. Machine is stopped for 47 minutes while maintenance replaces the motor.
Typical OEE impact: Equipment failure typically accounts for 5–15% of total scheduled time in poorly maintained facilities.
➜ Best countermeasure: Predictive maintenance — monitor motor current, vibration, temperature via OPC UA to identify degradation before failure. Moving from reactive to predictive maintenance reduces Loss 1 by 25–50% within 12 months.
What it is: Planned stops for product changeovers, tooling changes, calibrations, recipe adjustments, or cleaning procedures. These are expected stops but still represent time the machine isn't producing.
Manufacturing example: Changing from Product A to Product B requires 35 minutes: 20 min for tooling change, 10 min for parameter setup, 5 min for first-article verification.
➜ Best countermeasure: SMED (Single-Minute Exchange of Die) methodology — convert internal setup steps (done while machine is stopped) to external steps (done while machine is running). Target: reduce each changeover by 50%.
What it is: Brief interruptions (typically <10 minutes) that operators clear themselves — material jams, sensor trips, accumulation blockages, brief feed interruptions. These don't show up in maintenance downtime logs but accumulate significantly over a shift.
Manufacturing example: On a bottling line, label feed jams 12 times during a shift. Each jam takes 2–3 minutes to clear. Total: ~30 minutes lost, none of which appears in the maintenance downtime log.
➜ Best countermeasure: Video-based minor stop analysis — record the line for a full shift and categorise every stop. Even resolving the top-2 jam causes typically recovers 3–8 OEE points.
What it is: Running the machine below its designed ideal cycle time. Causes: operator caution with a new product, aging drive components running below spec, material quality variation requiring slower processing, undocumented "safe speed" culture.
Manufacturing example: A press has a nameplate cycle time of 0.8 sec/part. Operators run it at 1.1 sec/part "to be safe" — a 37.5% speed loss that shows up as a Performance shortfall even though the machine never stops.
➜ Best countermeasure: Display actual vs. ideal cycle time on the Andon board in real time. Operators who can see the gap immediately tend to investigate the cause. Investigate why the "safe speed" culture exists — often it's a legitimate material or machine issue that needs fixing at the root.
What it is: Scrap and rework produced during stable production. The machine is running at speed, but a fraction of output fails quality inspection — dimensional nonconformance, surface defects, functional failures.
Manufacturing example: During a 6-hour production run, 2.3% of machined parts fail dimensional inspection — typically when the cutting tool has accumulated wear past the half-life point.
➜ Best countermeasure: Statistical Process Control (SPC) on process parameters. Track tool wear, pressures, temperatures against control limits. Act when approaching the upper control limit — before defects appear.
What it is: Non-conforming parts produced during machine startup, after a changeover, or after a fault restart — before the process reaches its stable state. Every changeover creates a transient period of startup losses.
Manufacturing example: After every product changeover, the first 15–20 parts produced are off-specification while process temperatures stabilize. At 4 changeovers per shift, this is 60–80 parts scrapped per shift to startup losses alone.
➜ Best countermeasure: First-article inspection programme — hold production pending first-article sign-off after every changeover. Separately, reduce changeover-to-stable-state time by improving process temperature management (pre-heating, faster parameter stabilisation).
Calculate your OEE component shortfalls:
Identify your Six Big Losses automatically → Shopfloor Copilot categorises every OEE loss in real time from OPC UA machine data — showing which loss type is costing you most today.
See Loss Analysis →