Sheet cutting optimization guide
How to turn rectangular sheet sizes and a cut list into a reproducible plan while accounting for kerf, trim, rotation and waste.
Start with the usable sheet, not the nominal sheet
A sheet optimizer should first define the rectangle that is actually available for parts. If a 2440 × 1220 mm panel needs a 10 mm trim on every edge, the usable region is smaller than the purchased sheet. Kerf then consumes additional material as parts are separated.
PackMize keeps these inputs explicit so a user can reproduce the same plan later. PackMize does not infer machine allowances or material defects automatically.
1. Check the area lower bound
Area cannot tell you the layout, but it is a useful sanity check. Suppose the cut list is:
| Piece | Quantity | Area each | Total area |
|---|---|---|---|
| 800 × 400 mm | 6 | 320,000 mm² | 1,920,000 mm² |
| 600 × 300 mm | 4 | 180,000 mm² | 720,000 mm² |
| 250 × 200 mm | 8 | 50,000 mm² | 400,000 mm² |
Total requested part area is 3,040,000 mm². A 2440 × 1220 sheet has 2,976,800 mm² of nominal area, so one sheet is impossible even before kerf and trim. The mathematical lower bound is therefore at least two sheets.
That does not prove two sheets are sufficient. Geometry may force a third sheet if the rectangles cannot be arranged efficiently.
2. Model kerf before trusting a layout
Kerf is the width of material removed by a saw blade, laser, router or other cutting process. A zero-kerf diagram can show parts touching edge-to-edge even though the real machine needs space between cuts. PackMize uses the entered kerf as a conservative spacing allowance in its placement model. The separation allowance is required between parts and clipped at the usable outer edge. A full-width finished part can therefore use the existing sheet edge; enter edge trim when that edge needs cleaning.
3. Decide whether rotation is physically allowed
Allowing a 90-degree rotation increases the search space and can reduce waste. It is appropriate for many isotropic sheets and non-directional parts. It may be wrong for timber grain, decorative veneers, printed graphics, brushed metal, directional composites or parts whose machining orientation matters.
When rotation is disabled, PackMize preserves the entered width-height orientation for every piece. When it is enabled, the optimizer can also evaluate the 90-degree alternative.
4. Why one sorting rule is not enough
Two-dimensional cutting is a combinatorial packing problem. A common strategy is to place the largest parts first, but “largest” can mean area, longest side, perimeter or another measure. A sequence that works well for one cut list can trap unusable gaps in another.
General nesting compares 35 deterministic combinations of piece ordering and MaxRects-style placement scores. It prefers solutions with fewer unplaced pieces and fewer sheets. When the sheet count reaches a valid mathematical lower bound it is marked PROVEN OPTIMAL; otherwise it remains BEST SOLUTION FOUND.
5. Interpret utilization and waste
Material utilization is calculated from useful requested piece area relative to the full area of the sheets used. Material outside finished parts includes offcuts, kerf allowance and edge trim, shown separately. Eligible offcuts must meet the minimum-side threshold on both dimensions. That makes the metric useful for comparing plans, but it is not the same as recoverable scrap value. A large offcut may be reusable while many small gaps are not.
| Metric | Meaning |
|---|---|
| Sheets required | Number of stock sheets used by the selected plan |
| Pieces placed | Requested rectangles successfully assigned |
| Utilization | Useful requested part area ÷ total sheet area used |
| Waste | Sheet area not represented by requested finished parts |
6. What a visual plan is good for
The diagram helps verify that the result is understandable: which sheet contains each piece, how parts are oriented and where unused regions remain. It is still a planning diagram, not a machine program. A production workflow may need cut sequencing, common-line cutting rules, tabs, lead-ins, tool diameter compensation or clamping logic that PackMize does not generate.
Cutting sequence: choose Guillotine mode for cuts across each current subpanel. The sequence follows edge trimming and is available in the preview and complete CSV. General nesting does not supply a guillotine guarantee.
Practical workflow
- Enter the real stock-sheet dimensions.
- Add edge trim if the outer sheet edges cannot be used.
- Enter a realistic kerf for the cutting process.
- Add piece dimensions and exact quantities.
- Disable rotation where grain or surface direction matters.
- Run the optimizer and inspect every sheet layout.
- Validate machine-specific sequencing and tolerances before production.
Frequently asked questions
What is kerf in sheet cutting?
Kerf is the material removed by the blade or cutting process. A cutting optimizer must reserve space for it or the theoretical layout can be impossible to manufacture.
Should I allow piece rotation?
Allow 90-degree rotation only when the material and part specification permit it. Disable rotation for grain, print, surface or directional constraints.
Does the PackMize sheet optimizer generate CNC code?
No. It creates rectangular planning layouts. It does not generate machine toolpaths, lead-ins, clamping strategy or irregular-shape nesting.
Is the lowest-waste layout guaranteed to be globally optimal?
Only when PackMize shows PROVEN OPTIMAL. The general mode evaluates 35 deterministic strategies, and guillotine mode evaluates 14 and proves the sheet count only when it reaches a valid theoretical lower bound; otherwise it reports BEST SOLUTION FOUND.
Algorithm reference: Jukka Jylänki’s RectangleBinPack research and MaxRects implementation describe heuristic rectangular packing. PackMize applies its own documented kerf and trim model; general nesting alone does not establish a guillotine cutting sequence; choose the dedicated mode for that requirement.