Overall Equipment Effectiveness (OEE) Calculation in Garments
Lesson Contents Overview
Learning Objectives
Comprehensive Technical GuidanceTextbook & Training SOP
Overall Equipment Effectiveness (OEE) Calculation in Garments
1. Why Garment Factories Need One Honest Number
Every garment factory already measures something: line efficiency, output per day, defect rate, downtime hours. The problem is that each measure on its own can look acceptable while the factory still misses shipment dates. A line can report 78 percent efficiency and still fail, because efficiency blends stoppages, slow running and defects into a single figure that hides which problem is actually hurting.
Overall Equipment Effectiveness, universally shortened to OEE, solves this by refusing to blend. It measures three separate losses and multiplies them into one number. If any one of the three is weak, the whole number falls, and the factory can see exactly where to act. That honesty is the entire point of OEE. It is not a report to impress a buyer. It is a diagnostic tool for the production floor.
OEE was developed in manufacturing, but it fits garment production well because sewing lines suffer the same three losses as any production system: machines stop, lines run below their ideal speed, and some pieces come out defective. The only garment-specific work is defining the terms carefully for sewing: what counts as planned production time, what the ideal cycle time is for a style, and what counts as a good piece. Get those three definitions right and OEE works. Get them wrong and OEE becomes another number nobody trusts.
2. The Three Factors, Defined for a Sewing Line
Availability answers one question: of the time the factory planned to produce, how much time was the line actually running? Planned production time is shift time minus planned breaks, planned maintenance and planned meetings. If a shift is 540 minutes with 60 minutes of planned breaks, planned production time is 480 minutes. Every unplanned stop, from a thread break to waiting for cut panels, is subtracted from that 480 to give run time. Availability is run time divided by planned production time.
Performance answers a different question: while the line was running, how fast did it run compared with its ideal speed? Ideal speed comes from the operation bulletin for the exact style on the line, not from another factory or from memory. If the ideal cycle time is 0.5 minutes per piece, then 420 minutes of run time should produce 840 pieces. If the line actually produced 720 pieces, performance is 720 divided by 840, which is 85.7 percent. A line with no stops at all can still score badly on performance if operators run slowly or the line is badly balanced.
Quality answers the third question: of everything produced, how much was good the first time? Only first-pass good pieces count. A piece that needed rework is a loss, even if it eventually shipped. If 720 pieces were produced and 36 were defective, 684 were good first time, and quality is 684 divided by 720, which is 95.0 percent. Counting reworked pieces as good output is the fastest way to destroy trust in OEE, because operators quickly learn that producing defects and fixing them looks the same as producing good pieces.
OEE itself is simply the three factors multiplied: availability multiplied by performance multiplied by quality. Multiplication matters. Each factor is a proportion of what survived the previous stage, so the result is always lower than any single factor. That is not a flaw. It is the reason OEE tells the truth when other measures flatter.
3. Worked Example: One Shift, Every Step Shown
Consider a sewing line in Gazipur running one shift. Planned production time is 480 minutes after breaks are excluded. Unplanned stops during the shift total 60 minutes: 25 minutes of thread breaks, 20 minutes waiting for cut panels, and 15 minutes for a machine breakdown. Run time is therefore 480 minus 60, which is 420 minutes, and availability is 420 divided by 480, which is 87.5 percent.
The line produces 720 pieces during its 420 minutes of run time. The ideal cycle time for this style is 0.5 minutes per piece, so ideal output in 420 minutes is 840 pieces. Performance is 720 divided by 840, which is 85.7 percent. Of the 720 pieces produced, 36 are defective and need rework, so 684 are good first time. Quality is 684 divided by 720, which is 95.0 percent.
OEE is 0.875 multiplied by 0.857142 (the unrounded 720 divided by 840) multiplied by 0.95, which is 0.7125, reported as 71.3 percent. Read the result the way a manager should: the largest loss is performance at 85.7 percent, followed by availability at 87.5 percent, with quality the strongest at 95.0 percent. The first improvement action therefore targets slow running, perhaps by rebalancing the line or retiming the bottleneck operation, not by buying new machines to reduce stoppages. Without OEE, the same factory would probably have attacked stoppages first, because stoppages are the most visible loss, even though they are not the largest one here.
A second, shorter check confirms the method. A line with 90 percent availability, 90 percent performance and 98 percent quality has an OEE of 0.90 multiplied by 0.90 multiplied by 0.98, which is 0.7938, or 79.4 percent. Three good-looking factors still combine into a result well below any single one. Managers who are surprised by this are exactly the managers who need OEE, because they have been reading the factors separately and assuming the whole is as good as the best part.
4. OEE for a Single Bottleneck Machine
Line OEE is useful, but the most valuable OEE in a garment factory is often machine OEE on the bottleneck. A buttonholing machine, a pocket setter or a profile stitcher can hold up an entire line, and its losses deserve separate attention.
The method is identical. Take a buttonholing machine with 480 minutes of planned production time. It stops for 40 minutes during the shift, so run time is 440 minutes and availability is 440 divided by 480, which is 91.7 percent. Its ideal rate is 3 pieces per minute, so ideal output in 440 minutes is 1,320 pieces. It actually produces 1,150 pieces, so performance is 1,150 divided by 1,320, which is 87.1 percent. Of those, 23 are defective, so 1,127 are good, and quality is 1,127 divided by 1,150, which is 98.0 percent. Machine OEE is 0.917 multiplied by 0.871 multiplied by 0.98, which is 0.7827, or 78.3 percent. The bottleneck here is performance, and the fix is probably operator training or a jig change, not a maintenance programme.
Machine OEE also settles arguments about new equipment. If a bottleneck machine runs at 78.3 percent OEE and the line still misses targets, the factory can calculate exactly how much capacity a second machine or a faster machine would add, instead of buying equipment on instinct.
5. Setting Up OEE Without Drowning in Data
The most common reason OEE programmes fail is over-ambition at the start. A factory tries to measure every line, every machine and every loss from day one, the data collection collapses within a week, and OEE is declared impractical. The correct start is deliberately small.
Choose one bottleneck line. Agree the three definitions in writing: planned production time, ideal cycle time from the operation bulletin, and first-pass good as the only quality that counts. Give the supervisor one sheet per shift with columns for stop start time, stop end time, reason in plain words, and pieces produced per hour. Record every stop, including stops under five minutes, because ten short stops often cost more than one long breakdown, and short stops are the loss factories most often ignore.
Collect data for two full weeks before drawing conclusions. One day proves nothing: a bad day may reflect one breakdown, and a good day may reflect an easy style. Two weeks of daily OEE, with the three factors shown separately, reveal the pattern. Then rank the losses by total minutes or total pieces across the two weeks and attack the largest one first. The ranking step is what turns OEE from a measurement into an improvement programme.
6. The Weekly Review That Makes OEE Work
Measurement without review is decoration. The weekly OEE review should take 30 minutes and follow a fixed agenda. First, show the three factors for each day of the week, not just the average, so a bad day cannot hide inside a weekly average. Second, name the largest loss of the week in minutes or pieces. Third, assign one owner and one deadline for the action against that loss. Fourth, check last weeks action: did it move the number? If it did not, the action was wrong or incomplete, and saying so honestly is part of the routine.
Targets deserve care. A widely quoted benchmark from manufacturing literature is 85 percent OEE, built from 90 percent availability, 95 percent performance and 99.9 percent quality, but few garment lines reach it, and setting it as an immediate target only encourages data manipulation. A realistic path is to measure honestly first, then improve the largest loss by a few points each quarter. A line moving from 67 percent to 76 percent in a quarter has achieved something real. A line claiming 90 percent from week one has usually achieved creative accounting.
One rule protects the whole system: never punish a line for an honest low OEE in the first month. The first month exists to find the truth. If operators learn that honest data brings punishment, they will learn to hide stops, and OEE will become the prettiest useless report in the factory. Reward honesty first, improvement second, and the numbers will follow.
7. Common Traps and How to Avoid Them
The first trap is using total shift time instead of planned production time, which flatters availability by counting breaks as production. The second is counting reworked pieces as good, which hides quality losses. The third is using an ideal cycle time copied from another factory or an old style, which makes performance meaningless. The fourth is excluding short stops, which hides the largest loss on most lines. The fifth is publishing OEE without naming an owner per loss, which guarantees the number is admired weekly and never improves.
A subtler trap is comparing OEE across different styles without adjustment. A complex jacket line and a basic T-shirt line have different ideal cycle times and different defect profiles, and their OEE figures are not directly comparable. Compare a line against its own history on similar styles, and compare factories only after confirming identical definitions. OEE is a powerful internal improvement tool and a poor league table unless definitions match exactly.
When OEE is run honestly, it changes the conversation on the factory floor. Arguments about whether the line is performing well end, because the three factors show exactly where the loss sits. Supervisors stop defending averages and start fixing the largest loss. That shift, from defending numbers to fixing losses, is the real value of OEE, and it is worth more than any single percentage point the metric itself reports.
1. OEE Factors and Their Garment-Factory Definitions
| Factor | Formula | Worked Example | Result |
|---|---|---|---|
| Availability | Run time / Planned production time | 420 / 480 minutes | 87.5% |
| Performance | Actual output / Ideal output in run time | 720 / 840 pieces | 85.7% |
| Quality | First-pass good pieces / Total pieces | 684 / 720 pieces | 95.0% |
| OEE | Availability x Performance x Quality | 0.875 x 0.857142 x 0.95 | 71.3% |
2. Common Sewing-Line Losses by OEE Factor
| Factor | Typical Losses | First Fix to Try |
|---|---|---|
| Availability | Thread breaks, needle breaks, machine breakdowns, waiting for cut panels | Preventive maintenance schedule and a bundle buffer before the line |
| Performance | Unbalanced operations, slow running on difficult operations, operator absence cover | Rebalance the line and retime the bottleneck operation |
| Quality | Puckering, skipped stitches, wrong measurements, oil stains | Root-cause the top defect and retrain on that one operation |
Practical Commercial Flow
A sewing line in Gazipur runs one shift with 480 minutes of planned production time after breaks are excluded. Unplanned stops total 60 minutes, so run time is 420 minutes. Availability is 420 divided by 480, which is 87.5 percent. The line produces 720 pieces in its run time. The ideal cycle time is 0.5 minutes per piece, so ideal output in 420 minutes would be 840 pieces, and performance is 720 divided by 840, which is 85.7 percent. Of the 720 pieces, 36 are defective, so 684 are good first time, and quality is 684 divided by 720, which is 95.0 percent. OEE is 0.875 multiplied by 0.857142 (the unrounded 720 divided by 840) multiplied by 0.95, which is 0.7125, or 71.3 percent. The largest loss is performance, so the first improvement action targets slow running, not stoppages.
Factory Real-Life Scenario
Scenario: A woven shirt factory in Gazipur reported line efficiency of 78 percent but kept missing shipment dates, and management could not explain why. The IE team suspected the efficiency figure was hiding the real losses.
Solution: The team measured OEE on the worst-performing line for two weeks. Availability was 82 percent because of frequent thread-break stops on old single-needle machines, performance was 88 percent because operators slowed down on a difficult placket operation, and quality was 93 percent because of puckering on lightweight fabric. OEE was 67.1 percent, far below the reassuring efficiency figure.
Impact: Management replaced worn feed dogs and retimed the placket operation, availability rose to 89 percent and performance to 92 percent, and OEE climbed to 76.2 percent within a month. Shipment dates were met for the first time that season.
Commercial Officer Best Practice
Record losses in the operators own words first, then code them later. A code list written by management alone always misses the real reason the line stopped.
Top Beginner Mistakes to Avoid
- Using total shift time instead of planned production time as the availability denominator, which flatters availability by counting breaks as production.
- Counting reworked pieces as good pieces, which hides quality losses inside a respectable OEE figure.
- Using an ideal cycle time copied from another factory instead of the approved operation bulletin for that exact style.
- Excluding short stops from the record, even though short stops are usually the largest hidden loss on a sewing line.
- Publishing OEE without naming an owner for each loss, so the number is admired every week and never improves.
Important Commercial & Trade Terminology (5 Terms)
A single metric equal to availability multiplied by performance multiplied by quality, showing how much of planned production time actually produced good pieces.
💡 Note: The standard way to compare lines, shifts and factories honestly.Run time divided by planned production time, showing how much planned time the equipment was actually running.
💡 Note: Planned breaks are excluded from planned production time.Actual output divided by the ideal output possible in run time, showing how fast the line ran while it was running.
💡 Note: A line can have zero stops and still score badly on performance.First-pass good pieces divided by total pieces produced, excluding anything reworked.
💡 Note: Rework is a loss, not output.Shift time minus planned breaks, planned maintenance and planned meetings.
💡 Note: The honest denominator for availability.Interview Preparation Questions & Answers (4 Questions)
Master these high-frequency interview questions for Commercial Officer, Merchandising, and Export Executive roles.
OEE equals availability multiplied by performance multiplied by quality. Multiplication is used because each factor is a proportion of what survived the previous stage: only run time can produce, only running speed produces pieces, and only good pieces count. Adding them would let a perfect quality score hide terrible availability.
Frequently Asked Questions (4 FAQs)
Clear, beginner-friendly answers to the most common questions about the Commercial Department.
Key Takeaways & Summary Points
Executive Lesson Summary
OEE turns three separate factory problems into one honest number. Availability exposes stoppages, performance exposes slow running, and quality exposes defects, and multiplying them prevents any single good factor from hiding the other two. Measured on one line, reviewed weekly, and tied to named actions, OEE becomes the fastest route from arguing about performance to actually improving it.