Cut List Optimizer

About this tool

Cutting a sheet of plywood down to a list of parts is a puzzle with an obvious cost and no obvious answer. Order too few sheets and you go back to the shop; order too many and you are storing offcuts you will never use. This tool places every part on a sheet, applies the width the blade takes, and tells you how many sheets the job really needs.

The layout is produced by a well-known family of heuristics: parts are placed largest first, each one goes into the position that leaves the least awkward remainder, and the leftover space is split with a straight cut each time. That is the same approach sheet-cutting software has used for years. It is not a proven optimum — deciding the perfect arrangement is a famously hard problem, and no online tool honestly claims otherwise — but it is deterministic: the same list always gives the same plan, so it can be checked, discussed and followed.

Two numbers decide how much you actually get out of a sheet, and both are easy to get wrong by hand. Kerf is the material the blade turns into sawdust on every cut; ignore it and the last strip on the sheet comes up short, which is the classic way a job stops halfway through. Rotation is whether a part may be turned ninety degrees. Turning it usually nests much better, so it is on by default — but a part with a visible grain, a veneer face or a pre-finished edge cannot be turned, and the layout has to respect that.

Why the same cut list always gives the same layout

Every rule and every tie-break is fixed, so the same sheet size, kerf and part list always produce the same plan. Nothing is random, and nothing depends on the order the parts happen to be typed in. That repeatability is what makes a plan checkable.

Parts go down largest first, because a big panel with nowhere to go is what forces an extra sheet, while offcuts can still be filled in later. Parts of equal area are ranked by longest side, then shortest side, then by their position in your list — never by their label.

That last detail matters more than it sounds. Sorting by the label text would order A1, A10, A11, A2, so the drawing would come out with what looks like scrambled numbering even though the layout itself was perfectly repeatable.

When two open spaces would leave equally short leftovers, the tool prefers the higher one, then the leftmost, then the orientation that needs no rotation. Taken together, these rules are the reason a plan can be verified: you can reproduce it, and so can anyone you hand it to.

Where the saw kerf is deducted from on each cut

Kerf is taken once per cut, from the offcut side only. Splitting a space with one cut leaves two remainders, and the blade width comes out of the material being cut away — never out of the part you asked for.

A cut removes material, so a part and the space it came from cannot both keep their full size. The tool holds the part at the dimension you entered and takes the kerf from what is left, which is the convention an actual saw obeys.

When a free rectangle is split, the cut runs along whichever side leaves the smaller remainder. That keeps the other leftover as one longer strip, and long strips are what later parts can still use; squaring off the wrong side turns usable stock into slivers.

Leftovers that come out at exactly zero are not created at all. Keeping them would put zero-width rectangles into the free list — spaces no part can ever fit into, which would quietly inflate the number of positions considered.

Why the yield figure can never pass 100 percent

Yield is the area of the parts actually placed divided by the area of the sheets used, so it stays at or below 100 percent even when some parts do not fit. Those parts are listed separately instead of inflating the numerator.

Two areas look interchangeable and are not: the total area on your list, and the area the layout managed to place. When a part is larger than a whole sheet it ends up in the does-not-fit list, and from that point the two numbers part company.

Dividing the requested area by the sheet area would return a yield above 100 percent in exactly that situation — a confident number describing a layout that was never produced. The figure on screen is built from placed parts, so it is always something you could actually achieve on the saw.

The complement is waste, and it is more than offcuts. It includes the kerf turned into sawdust, the margin left at the edges, and every part that found no place. A poor yield is a signal to try another sheet size or to allow rotation, not proof that the algorithm was unlucky.

The kerf eats 2–3% of the sheet

A typical circular-saw kerf is 3 mm. Rip a 2440 × 1220 sheet ten times and the kerf alone consumes 30 mm of width — a whole 30 mm strip, or 2.5% of the sheet.

Run the arithmetic: a cut along 2440 mm at 3 mm wide loses 2440 × 3 = 7320 mm². Ten cuts lose 73,200 mm², and a standard sheet is 2440 × 1220 = 2,972,800 mm² — so 2.5% of the board becomes sawdust. That is the same order of magnitude as the difference between an 85% and an 88% yield, which is why any estimate that omits the kerf will systematically overstate the usable area.

The relative cost grows with the number of cuts and shrinks with sheet size. At twenty cuts the loss climbs to roughly 5%; put the same twenty cuts on a half sheet (1220 × 1220) and the proportion doubles again. This explains a familiar frustration: the area calculation says "just enough", yet the material always comes up short. What is missing is usually not an error in the sum but the saw width.

That is why this tool deducts the kerf as a fixed width on every step rather than subtracting a percentage at the end. The difference matters: a percentage cannot know how many cuts a given layout makes, while a layout does. In practice the kerf is not constant either — a fresh blade or a slightly off-square cut turns 3 mm into 3.5 mm, so leaving 1–2 mm of slack is cheaper than a second trip for material.