Glass · Production
Glass cutting optimiser
Glass is not plywood, and that is why the free nesting tools get it wrong. A glass sheet is scored and snapped, so every cut has to run edge to edge: you take a strip off the sheet, then you cut pieces out of the strip. That single constraint changes the whole layout. This page nests your cutting list the way a cutting table actually works — strips, guillotine cuts, edge trim and breakout width — and tells you how many sheets you need, what yield you get, and what the offcut is costing you.
The plan is the shop-floor version of this: every piece labelled on the diagram, the cuts in the order the table makes them, and each offcut with its real size for the rack. It stays in your browser like everything else here.
Why glass needs its own optimiser
Search for a cutting optimiser and you will find a dozen free ones. They are all built for plywood, MDF, aluminium or vinyl, and they will happily hand you a layout with a small piece tucked into a corner. On a panel saw that is fine. On a glass table it is impossible, because glass is not cut — it is scored and snapped, and a snap runs the full length of whatever you are holding.
So a glass layout is built in strips. You take a strip off the full width of the sheet, snap it away, then cut your pieces out of that strip with cross cuts. Every cut is edge to edge on the piece it is dividing. That is what this page does, and it is why its answer is lower than a free panel nester's — and why the cutter will not throw it back at you.
The optimiser runs the list through several orderings — tallest first, widest first, largest area first, longest side first — and keeps the best result. It says which one won in the line of working underneath the layout, because on some lists the difference is a whole sheet.
What good looks like
| Yield | Waste | What it means |
|---|---|---|
| > 90 % | < 10 % | What optimisation software is sold on. Hard to beat by hand. |
| 80 – 90 % | 10 – 20 % | A good manual cutter on a friendly list. |
| 75 – 85 % | 15 – 25 % | The published industry range for shops cutting without optimisation. |
| < 75 % | > 25 % | Either the list is awkward against this sheet size, or the sheet is wrong for the job. |
The number that matters is not the percentage, it is the invoice. Multiply the waste by what you pay per square metre and then by the number of jobs you cut in a year. That is the figure to compare any tool against, including this one.
Worked example you can check by hand
The list this page opens with, on a 3210 × 2250 sheet.
The panel above is generated by the same code that draws the layout, so it cannot drift away from what the tool actually does. Change any input and it changes with it.
What this page does not do
- One stock size at a time. Real shops mix jumbo, DLF and offcuts from the rack, and choosing between them is a large part of the saving. This page nests onto one size and tells you the truth about that size — run it twice and compare.
- No offcut library. The leftover from this job is not carried into the next one, which is exactly what a shop does in practice and exactly where the real money is.
- Strips only, no rotated sub-strips. A production optimiser will turn a strip ninety degrees and cut it the other way when that pays. This one does not, so treat its yield as a floor, not a ceiling.
- Rotation is all or nothing. You cannot lock one piece and free another. On a mixed job with some coated and some clear, run the two lists separately.
- Rectangles only. No shapes, no cut-outs, no holes, no arrises, and no thought given to which pieces should be cut first so the table is not blocked.
- It does not know your glass. Coating side, tong marks, distortion direction, heat-soak batching and which rack the sheet is on are all decisions a cutter makes and this page cannot.