OEE Explained for Indian Factories (with a Rupee Example)

OEE (Overall Equipment Effectiveness) is a single score for how much good product a machine makes versus its theoretical maximum. It multiplies three ratios — Availability × Performance × Quality — so a line running 85% × 80% × 95% scores 65% OEE, meaning it loses about a third of its potential output. In rupees, that gap is real money every shift.

Most mid-size Indian plants — metal, textile, auto-components, pharma, FMCG — already sense their machines are "running slow," but few can put a number on it. OEE is that number. This guide defines it the way the international standard does, works a full example in INR for one machine, and shows why the biggest losses are the ones your operators never write down.

The formula: Availability × Performance × Quality

OEE is defined in the international standard ISO 22400-2:2014, which specifies each input precisely so plants can be compared on the same basis (ISO 22400-2:2014, iso.org). All three factors are ratios between 0 and 100%, and you multiply them:

• Availability — did the machine run when it was supposed to? Lost to breakdowns, changeovers/setups, and (very Indian) power cuts and material waits. Availability = Run Time ÷ Planned Production Time.
• Performance — when running, did it run at full speed? Lost to minor stops, small jams, and reduced speed. Performance = (Ideal Cycle Time × Total Count) ÷ Run Time.
• Quality — of what it made, how much was sellable first time? Lost to defects, rejects and rework. Quality = Good Count ÷ Total Count.

Because you multiply, the score punishes a weak link. Three "decent-looking" numbers — 85%, 80%, 95% — compound to just 65%. That is why a plant that feels busy can still be leaving a third of its capacity on the floor.

The Six Big Losses behind the three factors

Each factor maps to concrete losses — the Six Big Losses of Total Productive Maintenance (TPM), the loss taxonomy widely credited to Seiichi Nakajima (oee.com, Six Big Losses):

Availability and Quality losses leave paperwork — a breakdown gets logged, a reject gets binned. The Performance losses (minor stops and speed loss) are the invisible ones: a two-minute jam every twenty minutes, a machine dialled down 10% to "play safe." Nobody writes those down, yet across a shift they quietly eat the largest slice of many Indian plants' OEE.

A worked example — one machine, in rupees

Take one machine on a single 8-hour shift. The numbers below are indicative, chosen to be round; plug in your own.

Step 1 — Availability. Shift = 480 min. Planned breaks = 30 min, so Planned Production Time = 450 min. During the shift you lose 45 min to a changeover and 22 min to a power dip before the DG picks up = ~67 min down. Run Time = 383 min. Availability = 383 ÷ 450 = 85%

Step 2 — Performance. Ideal cycle time = 0.5 min/part (2 parts/min). Over 383 min of run time the machine should make 766 parts. It actually made 613. Performance = 613 ÷ 766 = 80%

Step 3 — Quality. Of those 613 parts, 31 were rejected or reworked; 582 were good. Quality = 582 ÷ 613 = 95%

OEE = 0.85 × 0.80 × 0.95 = 65%.

Now the rupees. At full effectiveness this machine's theoretical max over the 450-min planned time is 900 parts. At 65% OEE it delivers 582 good parts/shift. A world-class 85% line would deliver ~765 — a gap of about 180 good parts every shift.

Put a contribution margin on each part. For an auto-component or fabricated-metal part, assume an indicative ₹15/part margin:

One machine, one avoidable ₹15 margin, and you are looking at roughly ₹1.4 lakh a month — indicative, not a quote. Multiply by a line of ten machines and OEE stops being a maintenance metric and becomes a P&L line.

And the losses cost even when they don't stop you selling: a machine idling or running slow still burns sanctioned-load electricity and paid labour. Industrial energy charges run around ₹6.4–₹7.1 per kVAh for large industrial consumers in a state like Uttar Pradesh for FY 2026 (Mercom India, UP tariff order) — you pay that whether the machine is making good parts or spinning at 80% speed.

Where the hidden losses hide — and how cameras surface them

The example's weakest factor is Performance (80%) — the minor stops and speed loss that no shift book captures. This is exactly where most Indian plants' OEE data is guesswork: Availability and Quality are logged, but the Performance bucket is estimated, so the improvement effort aims at the wrong loss.

You can close that blind spot without wiring every machine into a PLC or MES. A camera already pointed at the cell can tell running from stopped and timestamp every transition — turning "the line felt slow today" into a minute-by-minute record of which cell stopped, how often, and for how long. That converts the invisible Performance losses into a Pareto you can act on. This continuous machine-state and floor record also supports the operational discipline you already carry under the Factories Act, 1948 (e.g., §21 fencing of dangerous machinery, §22 work on or near machinery in motion) (India Code, Factories Act 1948) — the same feed that flags a stopped machine flags an unguarded one.

Knowing which cameras, watching which machines, from which angle is its own problem. That is the gap Mama is built to close: you record a short phone walkthrough of the floor, and it returns a floor plan plus a camera-placement plan — including which machines to watch for downtime and where the sightline to a machine's running-state indicator is currently blocked. You get the layout and the monitoring plan in a day, without waiting on a site survey.

One caution before you point cameras at people: worker video is personal data under India's Digital Personal Data Protection Act, 2023 (India Code, DPDP Act 2023 (No. 22 of 2023)). Monitor machines for OEE with clear notice, a defined purpose and a retention limit — the same discipline you apply to biometric attendance.

How a plant head should start

• Measure one machine, one shift, by hand first — the arithmetic above takes 20 minutes and usually shocks people.
• Don't chase 85%. World-class 85% (≈90% Availability × 95% Performance × 99.9% Quality) is the top few percent of plants globally; most discrete manufacturers sit at 60–75%. Aim to move your number, not to hit a poster.
• Attack Performance losses first — they're the biggest and the least measured. That's the fastest rupee win.
• Instrument, don't estimate. Whether by PLC or camera-based machine-state monitoring, replace the guessed Performance number with a measured one before you spend on "fixes."

FAQ

What is a good OEE score for an Indian factory? Most discrete manufacturers worldwide run 60–75% OEE; "world-class" is ~85% and represents the top few percent, per widely cited TPM benchmarks. For a mid-size Indian plant, the useful target isn't a fixed number — it's a measured baseline plus steady improvement on your weakest of the three factors.

How do you calculate OEE simply? OEE = Availability × Performance × Quality. Availability = Run Time ÷ Planned Production Time; Performance = (Ideal Cycle Time × Total Count) ÷ Run Time; Quality = Good Count ÷ Total Count. Multiply the three ratios. The definitions are standardised in ISO 22400-2:2014.

Why is my OEE low even though the machines look busy? Usually because of Performance losses — minor stops and reduced speed — which rarely get logged. A machine can look "running" all shift yet operate at 80% speed with frequent short jams. Measuring running-vs-stopped state (via sensor or camera) exposes this hidden loss.

Is OEE the same as machine utilisation? No. Utilisation only asks whether the machine was on. OEE also asks whether it ran at full speed (Performance) and made good parts first time (Quality). A machine can be 95% utilised and still score 60% OEE.

Can I measure OEE without an expensive MES or PLC integration? Partly, yes. Availability and Quality can start on paper. The hard part — continuous Performance/downtime data — can come from a camera watching machine state, which timestamps every stop without rewiring the machine. It's a practical starting point before a full MES project.