How a cut list optimizer works

The problem in one sentence

You have a list of parts to cut, each with a length, a width and a quantity. You buy material in full sheets. The job of a cut list optimizer is to arrange all those parts onto as few sheets (and your leftover offcuts) as possible, so you waste the least material, while making sure every cut is one you can actually make. That last part is where it gets interesting.

Kerf: the saw blade eats material

Every cut removes a thin strip of material, the width of the saw blade. That strip is called the kerf, and it is often around 3 mm. It sounds tiny, but across a sheet full of parts it adds up, and if the optimizer ignores it your parts come out slightly too small. A good optimizer subtracts the kerf at every cut, so the numbers on screen match the wood in your hands. In cutlistlab you set the kerf once and it applies to the whole sheet.

Guillotine cuts: why the layout has to be cuttable

A pretty layout is useless if you cannot cut it. A circular saw, a track saw and the panel saw at the hardware store all make guillotine cuts: a single straight cut that runs all the way across the piece, edge to edge. That means the layout has to break down into a series of full-width and full-height cuts, not a jigsaw of shapes that no saw can separate. A cut list optimizer that respects guillotine cuts gives you a plan you can follow in order, which is also why it can hand you a numbered cutting sequence to take to the workbench or the store.

Straight cuts or a CNC router

Guillotine cuts are a requirement of the circular saw, not of the tool. If you have a CNC router, it can also make internal cut-outs, so the layout is allowed to be tighter (non-guillotine). cutlistlab has a CNC mode (in Settings) that packs parts as closely as possible; the leftover material then often consolidates into one large usable offcut instead of scattered waste. Such a layout deliberately cannot be cut with a circular saw or rail: you export it as DXF (Export › CNC export) and send it to your router. Many tools show these non-cuttable layouts by default; we keep the regular calculation always straight-cuttable and put CNC mode aside as a deliberate choice. For most DIYers you leave the mode off and keep a plan you can put straight on the saw.

See the difference yourself: open the CNC puzzle, a pinwheel-style set of parts that CNC mode fits onto 2 sheets, while the regular saw mode needs 4. Toggle CNC mode on and off in Settings to compare the two layouts.

How the optimizer finds a good layout

Here is the honest part: packing rectangles onto sheets perfectly is one of those problems that gets impossibly slow to solve exactly once you have more than a handful of parts. So optimizers do not chase the perfect answer, they use strong rules of thumb (heuristics). cutlistlab tries several different packing strategies, for example sorting parts in different orders and placing them in different ways, runs each one, and keeps the best result. It then spends a little time trying to improve that result further. The outcome is near-optimal: not provably perfect, but consistently better than laying parts out by hand, and produced in a second or two.

Parts that have no grain requirement are free to rotate, which lets them slot into tighter gaps. Parts you mark as grain-fixed stay oriented the right way, even though that can cost a little efficiency, because matching grain matters more than squeezing out the last percent.

Verified optimal: the "Search thoroughly" button

Because the standard result is already near-optimal in a second, you can press Search thoroughly to have it verified. The tool then keeps trying thousands of extra layouts in the background using heavier search techniques (simulated annealing and a genetic algorithm on top of the regular strategies) and keeps the best. If it finds something better it adopts it automatically; if it finds nothing better after hundreds of attempts, you know your layout is as good as optimal. We show a theoretical lower bound alongside it: the minimum number of sheets you need based on area alone, so you can see how close you are to the mathematical minimum. In practice the regular calculation often reaches that minimum already, so the button mainly gives you certainty, honestly measured rather than an empty promise.

There is more than one "best": three goals

"Best" depends on what you are optimizing for, and they pull in different directions. cutlistlab lets you pick the goal, which genuinely changes the layout:

This is the part most tools skip. Optimizing purely for waste can give you a layout with dozens of fiddly cuts, which is no bargain if the store charges per cut or you are doing them all with a hand-held saw.

Offcuts come first

If you already have leftover panels, a good optimizer should use them before cutting into a fresh sheet. cutlistlab keeps an offcut and remnant stock across your projects and places parts on the smallest suitable piece first, so your pile of leftovers actually shrinks instead of growing.

Frequently asked questions