How does AI reduce fabric waste in marker making and cutting?
AI-assisted nesting can lift marker efficiency, but fabric constraints, patterns and order mix set the ceiling, and the savings need measuring.
KEY TAKEAWAYS Summary by the editors
- Marker making is the arrangement of pattern pieces on fabric before cutting, and efficiency is measured as cut area divided by fabric area, multiplied by 100.
- Automated nesting can be compared with manual markers and simulated against real yield, which accounts for lay thickness and cutting kerf.
- Constraints such as stripes and plaids, directional fabrics, defects and shrinkage limit how tightly pieces can be packed, so results differ greatly between fabrics.
- One cutting-room software supplier states that typical cutting-room waste is about 10 to 20%, a figure that comes from a vendor and should be checked against a mill's own baseline.
- McKinsey's Fashion on Climate analysis lists minimising production and manufacturing waste among the levers for manufacturers, but does not give cutting-specific yield figures.
How does AI reduce fabric waste in marker making?
AI reduces fabric waste by searching many more piece arrangements than a person can, subject to the constraints of the fabric and the order. The usual result is a marker with higher efficiency, meaning more of the fabric area becomes garment parts. The gain is real but bounded by the garment, the fabric and the size mix.
This explainer sets out what marker making is, how software changes it and what to measure before claiming a saving.
What is marker making and how is efficiency measured?
A marker is the layout of pattern pieces on a given width of fabric, used to guide cutting of many layers at once. Efficiency, often called yield, is calculated as cut area divided by fabric area, multiplied by 100. One source notes that a single percentage point of improved yield can save substantial material cost on large production runs, which is why small percentage differences matter in commercial terms.
Marker making has long been supported by CAD software. What AI adds is a more flexible search and the ability to learn from past results, for example which size combinations nest well together.
What does AI-assisted nesting do differently?
- It tests many arrangements and rotations of pieces within allowed rules, then selects the best result.
- It can compare automated markers with manual ones, so that a marker maker can accept, adjust or reject the proposal.
- It can simulate real yield, accounting for lay thickness and cutting kerf, before the fabric is cut.
- It can group sizes or styles that nest well together to reduce changeovers.
- It can export markers directly to cutting machines, reducing translation errors and rework.
A cutting-room software supplier's published guidance also lists digitising cut panels to correct distortions and moving away from paper as steps in reducing re-cuts. These are workflow improvements as much as algorithmic ones.
Which constraints limit how much waste can be removed?
| Constraint | Effect on nesting | What helps |
|---|---|---|
| Stripes and plaids | Pieces must match at defined windows, which leaves gaps | Specify match points early and check the first lay |
| Directional fabrics | Directional pieces cannot be rotated | Rotate non-directional parts only |
| Fabric defects | Critical parts must avoid holes and dye flaws | Track defect locations per roll |
| Usable width | Selvage reduces cuttable width and varies by roll | Record usable width per roll |
| Shrinkage | Pattern allowance must reflect fabric behaviour | Test and record shrinkage by lot |
| Order mix | Few sizes per marker limits packing options | Combine orders where lead times allow |
The practical consequence is that results transfer poorly between fabrics. A saving achieved on a plain, non-directional woven may not appear on a checked wool.
How much waste is typical, and what savings are realistic?
Published figures vary and many come from technology suppliers. One cutting-room software supplier states that typical cutting-room waste is about 10 to 20%, and presents AI-assisted nesting and digital checks as ways to reduce it. Treat such numbers as indicative, because they depend on the product mix and are not independent measurements.
McKinsey's Fashion on Climate analysis finds that manufacturers and fibre producers could deliver 61% of accelerated emissions abatement, including by minimising production and manufacturing waste. It does not provide cutting yield figures. A factory should therefore establish its own baseline before setting a target.
Data capture is often the hidden task. Marker software needs reliable fabric widths, shrinkage values and defect information, and many factories record these on paper or not at all. Capturing the data per roll, and linking it to the order and the marker used, also creates the evidence needed to report savings later. Without it, a factory can neither show that a new nesting approach worked nor identify the fabrics where results fall short.
Responsibility is shared between brands and factories. The brand sets pattern construction, grain rules and match requirements, which strongly influence achievable yield. A pattern with many directional pieces or demanding match points will always nest less tightly than a simple one. Design and technical teams can therefore contribute to waste reduction by reviewing patterns for nesting friendliness at the development stage, not only by asking the factory to improve afterwards.
Sustainability reporting is another reason to measure carefully. Fabric not turned into garments carries the emissions and water use of its production, so higher yield reduces the impact per garment. McKinsey's analysis counts minimising production and manufacturing waste among the levers open to manufacturers. To report a benefit, a team needs a baseline, a clear boundary, and consistent definitions of what counts as waste, including end-of-roll remnants and defective fabric.
How should a brand or factory introduce AI nesting?
- Measure the baseline by fabric type, including marker efficiency and re-cuts.
- Standardise fabric data per roll: usable width, shrinkage, pattern repeat and defect locations.
- Run automated markers beside manual ones on a sample of orders and compare yield.
- Check the first lay and pilot cut for pattern matching before releasing a full run.
- Track results by fabric group and keep manual rules for difficult cases.
Brands that do not cut their own goods depend on factories for this data. Asking suppliers to report marker efficiency by order, and agreeing how savings are shared, makes the improvement visible without forcing a particular tool.
What are the limits and risks?
Higher efficiency does not help if it increases defects, mismatched patterns or cutting time. Fabric savings should also be set against the cost and energy of the software and any extra handling. Finally, marker efficiency is one part of waste: overproduction and unsold stock usually account for more material than cutting scraps, so cutting improvements should sit alongside better demand planning.
Frequently asked questions
What is marker efficiency?
It is the share of fabric area that becomes garment pieces, calculated as cut area divided by fabric area, multiplied by 100. Higher efficiency means less fabric waste for the same garments.
How much fabric is wasted in garment cutting?
It depends on garment, fabric and order mix. One cutting-room software supplier states about 10 to 20% typical cutting-room waste, but this is a vendor figure. A factory should measure its own baseline.
Can AI nesting work on striped or checked fabrics?
It can, but stripes and plaids require pieces to match at defined windows, which limits packing. Gains are usually smaller than on plain, non-directional fabrics.
Does better marker making solve fashion's waste problem?
No. It reduces cutting waste, but overproduction and unsold stock are usually larger sources of waste. Cutting improvements work best alongside demand planning and design choices.
One edition every weekday morning. Read in five minutes. Free for industry professionals.
