DHU in garment factory work means defects per hundred units, and it is the number most quality teams put on the wall of the sewing line. It shows how many defects were found for every hundred garments checked. A line with a DHU of 3 finds three defects in every hundred pieces. A line with a DHU of 12 is making twelve, and each one costs time to repair or money to scrap.

The figure sounds simple, but many factories calculate it in different ways, mix it up with defective percent, or use it for the wrong purpose. This guide explains the formula, how DHU differs from defective percent and right first time, a worked example, where and how to collect the data, and how to bring DHU down in a way that lasts.
What DHU means
DHU stands for defects per hundred units. A defect is a single fault found on a garment, such as a skipped stitch, an open seam, a stain or a wrong measurement. A unit is one garment. DHU counts the defects, not the garments that carry them.
DHU = (total number of defects found ÷ total number of units inspected) × 100
One garment can carry more than one defect, so DHU can be higher than the percentage of garments that are defective. A garment with a skipped stitch, a stain and an uneven hem has three defects but is only one defective piece.
DHU, defective percent and RFT: three different numbers
| Measure | What it counts | Formula |
|---|---|---|
| DHU | Defects per hundred garments inspected | (defects ÷ units inspected) × 100 |
| Defective percent | Garments with at least one defect | (defective pieces ÷ units inspected) × 100 |
| RFT (right first time) | Garments passed with no defect at first check | (pieces passed first time ÷ units inspected) × 100 |
The difference between a defect and a defective is explained in our guide on defects vs defectives in garments. In short, a defect is the fault, and a defective is the garment that has it.
A worked example
An end line checker inspects 500 garments in a shift. The checker finds 35 defects in total. These sit on 28 garments, since some garments have more than one defect. The numbers are assumed for the example.
- DHU = (35 ÷ 500) × 100 = 7.0
- Defective percent = (28 ÷ 500) × 100 = 5.6 percent
- Pieces passed first time = 500 − 28 = 472
- RFT = (472 ÷ 500) × 100 = 94.4 percent
Each of the three numbers tells a different story. DHU shows how many faults the line is generating. Defective percent shows how many garments need repair. RFT shows how many garments went through the first check without any repair. A line can improve one without the others, so report all three.
Why DHU matters
- It shows the quality level of a line in one number, and the trend over days.
- It points to cost. Every defect means repair time, thread, handling and sometimes a rejected garment. See rework in a garment factory.
- It gives early warning. A DHU that rises through the day tells you to look at a machine, an operator or a lot of fabric before the whole shift is affected.
- It supports line comparison. Supervisors can see which lines are doing well and learn from them.
- It links to the buyer’s final result. Poor in line quality shows up as a failed final inspection. See the AQL sampling plan for garment inspection.
- It is a key figure in the factory scorecard. See KPI factors of garments.
Where DHU is measured
DHU can be measured at several points. Each gives a different view.
| Checkpoint | What it shows |
|---|---|
| Inline (at the operation or table) | Defects at the source, per operator and operation |
| End of line | The total quality of sewing before finishing |
| Pre final or audit | Quality after finishing, before the buyer’s check |
| Washing or special process | Damage caused in wet or special processes |
| Final inspection | The state of goods at packing |
Factories often track DHU at the end of the sewing line, since that is where every garment is looked at. Inline DHU is more useful for fixing causes, because the defect is found close to the operation that made it. Be clear which one you report, so that numbers are not mixed up.
How to collect DHU data
1. Define the defect list
Use one standard list of defects for the factory, with a code and a description, so everyone calls the same fault by the same name. A good list covers sewing defects, fabric defects, stains, measurement problems and finishing faults. See our guides on garments defects identification and sewing defects and root causes.
2. Decide what counts as a defect
State the rule for each defect, with photos if needed. A thread end of a given length may count, a shorter one may not. A checker who is unsure should follow the written rule, not personal opinion.
3. Count every defect
Each defect on a garment is recorded, even if there are several on the same piece. Do not stop at the first one.
4. Record by type, operation and operator
Write the defect code, the operation where it happened, and the operator if the line uses that level of tracking. Without this detail, DHU says only that there is a problem, not where.
5. Count the units checked
Record how many garments were inspected in the same period. If the checker only samples, state the sample size.
6. Calculate by hour and by shift
Compute DHU every hour on the line board, and the daily figure at the end of the shift. An hourly figure lets the supervisor act while the lot is still in progress.
7. Record repairs
Note which garments were repaired and passed at the second check, and which were rejected. This gives the real final loss.
Reading the data: Pareto and trends
Totals by defect type, shown from highest to lowest in a Pareto chart, usually show that a few defect types make most of the count. Work on these first. The same analysis by operation shows where the problem starts. Our guide on 7 QC tools in the apparel industry shows how to use the Pareto chart, the check sheet and the cause and effect diagram.
Look at trends too.
- By hour: a rise after lunch may point to fatigue or a machine problem.
- By day: a jump on a new style shows a method or training problem.
- By lot: a spike on one fabric roll points to a fabric issue.
- By operator: a pattern may call for training or a better method, not blame.
What is a good DHU?
There is no single standard. The target depends on the buyer, the style, the fabric, the process and the factory. Simple styles in a stable line can reach a low DHU. A complicated style, a new style or a new operator group will start higher and fall as the line settles. Many factories set a target with the buyer or from their own best performance, and track against it. Do not compare DHU between very different styles without allowing for the difficulty.
Also be careful about targets that push staff to hide defects. A DHU that is low because defects are not counted is worse than a high one that is true. Reward honest recording and improvement, not just a low figure.
How to reduce DHU
Find the cause, not only the defect
Repairing defects does not reduce them. The aim is to stop them being made. For each of the top defects, ask why it happens, and keep asking until you reach a cause you can fix.
| Common cause group | Examples |
|---|---|
| Machine | Wrong needle, wrong tension, worn parts, poor maintenance |
| Method | No standard, no guide, poor sequence of work |
| Material | Fabric faults, shade variation, wrong trims, shrinkage |
| Man | Low skill, new operators, fatigue, lack of training |
| Measurement | Poor patterns, wrong marker, wrong gauge |
| Environment | Poor light, dust, heat, noise |
Actions that work
- Train the operator on the defect. Show the defect, the standard and the correct method. See sewing operator training program.
- Check the first piece of each operation, after each change of style or machine setting, before starting the run.
- Inspect early. Inline checks catch problems at the first operation instead of at the end of the line.
- Set up the machine properly. Needle, thread, tension, stitch length and feed should match the fabric. Keep a maintenance plan. See sewing machine maintenance schedule.
- Use guides, folders and templates so the operator does not depend on skill alone.
- Design out the error. Make it hard to do the work wrong. See poka yoke examples in garment manufacturing.
- Improve lighting and workplace. See lighting in a garment factory.
- Control the material. Inspect fabric and trims on arrival. Rest knit fabric before cutting. See knit fabric relaxation before cutting.
- Give feedback fast. Show the defect to the operator the same hour, with the garment, not at the end of the week.
- Use small team improvement. A group of operators and a checker can solve a recurring defect together. See quality circle in the garment industry.
- Hold a daily review. Look at the DHU board, the top three defects, and the actions agreed.
Link DHU with the system
DHU should link to the quality control system and its documents. Our guides on the sewing quality control SOP and quality control in garment manufacturing show how checks, records and responsibilities fit together.
Common mistakes in DHU reporting
- Counting defective pieces and calling it DHU. The two numbers differ.
- Counting only the first defect on a piece.
- Using different defect lists on different lines, so the numbers cannot be compared.
- Changing the number of pieces checked, which changes the result. Say if it is a sample or a full check.
- Reporting only a daily figure, so no action is possible during the shift.
- Not recording the operation, so the cause is unknown.
- Pressure to show a low number, which leads to hidden defects.
- Not linking to action. A board full of numbers with no follow up changes nothing.
- Ignoring repaired defects. A defect that is repaired still happened, and it still counts.
DHU on the line board
A simple board near the line makes DHU visible. Show these items.
- The target and the actual DHU by hour
- The top three defects of the day, with counts
- The operations where they happen
- The actions agreed and the person responsible
- RFT and defective percent for the day
Keep the board current, in clear numbers, and review it with the line team at set times. Where possible, collect data through a tracking system rather than by paper. See real time production tracking on the sewing floor.
Final thoughts
DHU in garment factory work is the number of defects found per hundred garments inspected, calculated as defects divided by units inspected, times one hundred. It is not the same as defective percent, which counts garments, or RFT, which counts garments that passed first time. Report all three, and collect the data by defect type and operation.
The value of DHU lies in the action it triggers. Use it hourly, find the top defects, fix their causes with training, method and machine care, and reward honest numbers. A DHU that falls because the causes were removed is worth far more than one that falls because the checking was relaxed.
Frequently asked questions
DHU means defects per hundred units. It tells how many defects were found for every hundred garments inspected.
DHU = (total defects found ÷ total units inspected) × 100.
DHU counts every defect, so one garment with three faults adds three. Defective percent counts the garments that have at least one defect.
Right first time is the percentage of garments that pass the first inspection with no defect and no repair.
There is no single standard. It depends on the style, buyer and factory. Set a target for each style and track the trend.
At inline checkpoints for finding causes, and at the end of the line for the total sewing quality. Always say which one you report.
Find the top defects with a Pareto chart, remove their causes through training, machine setup, method and material control, check early, and give quick feedback to operators.
