Field note · Craft
A fabric nesting result is a good candidate layout, not a proof of global optimum
Published: September 9, 2026 · Maintained by: Jisung Kim · AI assistance and release-check practices are disclosed in the editorial policy. This note documents WEBBE-B tool behavior, design choices and verification practice.
Fitting required pieces into leftover fabric is a geometric packing problem with practical constraints. ScrapFit searches several layouts and scores them, but it deliberately does not claim that the result is the best arrangement that could exist. That distinction is important because exact global optimisation can become computationally expensive as pieces and irregular boundaries multiply.
Piece order changes the outcome
If a large rectangle is placed first, it may reserve the only region where it fits and leave small pieces to fill gaps. If small pieces are placed first, they can fragment the remnant and block the large piece. Different sorting strategies therefore produce different valid layouts from the same inputs.
Rotation is not always free
A paper rectangle can often rotate 90 degrees without consequence. Fabric may have grain, directional print, stretch or nap. A piece that geometrically fits after rotation may be unusable in the real project. ScrapFit lets rotation and grain constraints participate in the search rather than treating every rectangle as interchangeable.
Defects turn a rectangle into a constrained region
Stains, tears, holes and already-used zones remove areas from the remnant. A layout that only compares outer width and height can place a required piece directly over a defect. Modelling defect zones makes the problem closer to the physical scrap the user actually has.
Irregular remnants need approximation
An L-shaped remnant or simple polygon cannot be handled by the same rule as one perfect rectangle. Candidate positions need containment checks against the actual boundary. Curved edges can be approximated by multiple straight segments, but the user should leave a safety margin because the digital polygon is still an abstraction of a flexible material.
Grid precision trades time for search density
A browser can search candidate positions on a grid. A smaller step checks more placements and can find tighter fits, while a larger step is faster. The important design choice is to report that precision so the user knows how densely the space was sampled. “No fit found” at a coarse grid does not prove that no physical arrangement exists.
A worked remnant example
Suppose there are three remnants, eight patch pieces and one damaged rectangle. A large-piece-first strategy completes seven pieces with low waste, while a randomized order completes all eight but scatters leftovers into smaller regions. Which is better depends on the objective: completing the current project, preserving one large future scrap, or minimizing total waste. A single “best” score can hide those different goals.
Why “try another layout” is a feature
Heuristic search benefits from alternative candidates. Showing another valid arrangement gives the user evidence about how sensitive the plan is to order and scoring. If several searches converge on similar placements, confidence increases. If they differ dramatically, the user may want to compare them before cutting.
Physical cutting adds constraints the optimiser does not measure
Seam allowance, blade width, fraying, stretch, print registration and marking accuracy can all reduce usable material. Digital placements should include spacing that reflects the chosen process. Structural or safety-critical textile work can require professional rules beyond a generic nesting tool.
The correct claim is practical, not perfect
ScrapFit is valuable when it finds a workable arrangement faster than manual trial and error and makes its assumptions visible. It is not a mathematical certificate of global optimality. That honest boundary lets the user combine computation with the physical knowledge the browser does not have.
Compare layouts by the same objective
Two nesting runs are only meaningfully comparable if they use the same fabric width, piece dimensions, rotation rules, spacing allowance and defect exclusions. Changing one constraint can make a lower-area layout look better while actually violating the production requirement. A useful comparison records both the utilization percentage and the remaining continuous offcut shapes, because a layout with slightly lower utilization may leave a reusable rectangle while a denser layout leaves only narrow scraps. The “best” candidate therefore depends on the next job, not one percentage alone.
Print a scale check before cutting
When a digital layout is transferred to paper or a projector, confirm one known dimension with a ruler before using it as a cutting guide. Browser zoom, printer scaling and PDF fit-to-page settings can all change physical scale even when the nesting math itself is correct.