Methodology
How the pattern is computed.
Komon Lab is, at heart, a translation of traditional wasai proportional rules into a small set of formulas. This page documents those formulas, the sources behind them, the known limits, and the version history.
Why this page exists
Any pattern generator is only as good as the rules inside it. If the rules are wrong, the pattern is wrong in a way that is hard to spot from looking at the PDF. So Komon Lab publishes its rules — the same way a research paper publishes its methods — and asks anyone with sewing experience to disagree.
The intent is twofold. For sewers, this page is a reading guide to what assumptions are baked into your printed pattern. For other practitioners — independent pattern makers, wasai teachers, sewing-history researchers — it is an invitation to point out where the rules are wrong, oversimplified, or out of date.
The seven inputs
The tool asks for seven body measurements. They are described in the measuring guide in plain English. Here we give them their conventional wasai names and the symbols we use in the formulas below.
- 身長 (shinchō, height,
H): floor to crown of head. - 裄 (yuki, sleeve reach,
Y): from centre-back neck to wrist with the arm at 45°. - 身幅 reference (mihaba reference,
M): half-hip plus ease, used to size body panels. - 裄丈 reference (yukitake,
YT): finished kimono length, typically floor to ankle for everyday wear. - 袖丈 (sodetake, sleeve drop,
SD): shoulder seam to chosen sleeve hem, varies by formality. - 肩幅 (katahaba, shoulder half-width,
KH): centre-back to shoulder point. - 胸囲 (kyōi, bust circumference,
B): used as a cross-check and for the overlap allowance.
A simplified set of seven measurements covers the great majority of straight-cut everyday kimono. Specialised garments — formal furisode, men's haori-hakama sets, ceremonial layers — would need more. The current tool intentionally aims at the most-used case first.
Derived dimensions
From the seven inputs, the tool derives every panel dimension the cut chart needs. The formulas below use the symbols introduced above. They are stated in centimetres, but the tool itself accepts inches and converts internally.
Body length (mitake, L)
The finished length from shoulder seam to hem. For everyday women's kimono, L = YT + 0 (the input is already finished length). For formal women's wear with a built-in ohashori tuck, the unfolded body length is L = YT + (1.10 · H − YT), where the extra material is taken up by the tuck. Men's kimono have no ohashori and use L = YT. The tool currently implements the everyday women's case and the men's case; the formal-tuck case is on the roadmap.
Sleeve reach to sleeve panel width (yuki → sodehaba)
The sleeve reach Y is split between the body panel half-width (katahaba) and the sleeve panel width (sodehaba), Y = KH + sodehaba. The body panel half-width is read directly from the input KH; the sleeve panel width is computed as sodehaba = Y − KH. The conventional rule is that sodehaba should not be less than 30 cm even on small frames, because below that figure the sleeve becomes too narrow to fold cleanly. The tool clamps to a 30 cm floor and warns if the input combination forces it.
Body panel width (migoro haba)
Each of the two body panels is cut to the bolt width if cutting from tan-mono, or to half the bolt width plus seam allowance if cutting from a 110–150 cm Western bolt. The proportional rule used internally is migoro_haba = M · 0.5 + ease, where ease defaults to 4 cm for women and 6 cm for men. The ease number is taken from contemporary wasai teaching practice rather than from a frozen historical figure, because modern sewers wear modern undergarments and the historical figures assume a different layered baseline.
Okumi (overlap panel) width
The okumi is the diagonal panel that gives a kimono its wrap-front silhouette. The conventional okumi width is 15 cm at the hem, tapering to roughly 4 cm at the upper neckline. The tool implements okumi_lower = 15 cm fixed, okumi_upper = 4 cm fixed, and computes the angle as a function of the body length L. The 15 / 4 numbers are from standard women's everyday wasai; the men's variant uses slightly narrower 13 / 3.5 cm by tradition, and the tool follows that when the male preset is selected.
Collar (eri) length and width
The collar is cut as a long rectangle, length eri_length = L + 2 · sodetake (roughly: enough to run down the front, around the neck, and down the other front to the same depth as the sleeves). The collar width is eri_haba = 5.5 cm for women's everyday wear, doubled before sewing for the kakeeri overlay. Men's wear uses a 5 cm width by tradition. These figures come from the same source pool described in the sources section below; small variations of half a centimetre exist in the literature.
Sleeve panel (sode) shape
The sleeve is a rectangle sodehaba × (2 · sodetake), folded in half along the long axis to give the finished sleeve. The corner near the wrist is curved on women's wear (the maru-sode curve), with a curve radius equal to roughly half the sleeve drop, typically 12–18 cm. Men's wear uses a square corner. The tool implements both and exposes the choice in the input form.
Seam allowances
Traditional wasai uses generous seam allowances — 2 cm on most seams, 4 cm on the hem and sleeve hem — so the garment can be unstitched, washed, and re-stitched at a slightly different size later in its life. The tool defaults to these traditional values. A "modern" preset is available that uses 1 cm seam allowances throughout, which makes the garment easier to construct but harder to alter or rebuild.
Cut chart logic
Once the panel dimensions are known, the cut chart arranges them on the fabric width the sewer has selected. The logic, in plain language:
If the fabric width is between 33 and 40 cm (a traditional tan-mono), the panels are laid in a single column. The total fabric length is the sum of panel lengths plus seam allowances, plus a 10 cm safety margin. This is the historically standard layout.
If the fabric width is between 100 and 120 cm (Western quilting-cotton range), the body and lining panels are paired in two columns; the sleeve panels share a third row; the collar runs along the selvage. The yardage figure is the longest column plus margin.
If the fabric width is between 130 and 160 cm (fashion-fabric range), the layout adds a fourth column for the okumi panels and reduces yardage by roughly 20 percent compared to the 110 cm case. This is the most efficient layout for the typical Western wardrobe fabric.
The tool will refuse to produce a cut chart for fabric widths below 30 cm or above 170 cm, because the layout assumptions break down at the extremes. The fabric widths guide gives the longer reasoning.
Sources behind the formulas
The proportional rules above are drawn from the following bodies of literature, in rough order of how much weight each carries in the current tool:
Contemporary Japanese wasai curricula. The proportional ratios for body width, sleeve width, and overlap are taken from the standard teaching practice in modern wasai schools. These figures are remarkably stable across schools — most agree to within half a centimetre — and represent the working consensus of practising tailors. Where individual schools disagree, the tool sides with the majority and notes the disagreement here.
John Marshall, Make Your Own Japanese Clothes (1988). One of the most thorough English-language treatments of wasai. Marshall's measurement system is broadly compatible with ours; the tool's default ease numbers are within 0.5 cm of his.
Liza Crihfield Dalby, Kimono: Fashioning Culture (1993). Used primarily for context on formality ranks and historical proportion shifts. Dalby is descriptive rather than prescriptive, and the tool draws on her work to choose which proportional rules to implement first.
Sharon Sadako Takeda and Luke Roberts, Japanese Fishermen's Coats from Awaji Island (2001). Used for cross-checking the men's preset and the traditional seam-allowance figures.
Mary Dusenbury and Carol Bier, Flowers, Dragons and Pine Trees (2004). Used for context on regional weaving practice and the proportions of antique garments in museum collections.
A fuller reading list is included with most of the long-form guides on this site. The methodology page lists only the sources actually consulted while writing the formulas; the per-guide source lists are wider.
Known limits
The current tool has several known limits. We list them here so readers can plan around them.
Unusual body proportions. The proportional rules assume a relatively standard human form — bust, waist, and hip within about 25 cm of each other, height between 145 and 195 cm. Bodies outside that range will get a pattern that works but may need manual adjustment, particularly at the bust and hip. We do not believe the rules become wrong; we believe the ease defaults become too generous or too tight.
Posture differences. A kimono with a built-in ohashori tuck assumes a specific posture — relaxed shoulders, neutral spine — and the tuck reads as messy when the wearer has marked forward-shoulder posture. The current tool does not adjust for posture; this is on the roadmap.
Antique fabric variance. Tan-mono bolt widths in antique fabric vary more widely than modern bolts. The tool's 33–40 cm range covers most modern tan-mono, but if you are working from an antique bolt narrower than 32 cm — common in 19th-century rural weaving — the layout assumptions break. Hand-layout is the only safe option in that case.
Children's age-fold. The traditional age (a sewn tuck that lets a child's kimono grow with them) is not yet implemented. A children's preset that includes the age-fold is planned for a future version. Our guide to the age-fold describes the principle.
Formal and ceremonial garments. The tool currently targets komon-tier and tsumugi-tier everyday wear, plus yukata. It does not output the proportions for furisode, uchikake, tomesode, or other formal-rank garments. This is deliberate — those garments carry meaning that a generated pattern would risk distorting — and there is no current plan to add them.
Version history
The generator is versioned. Older versions are reproducible: if you produced a pattern on v1.2 and want to regenerate the same pattern on v2.0, the version selector in the tool lets you pin to a specific generator version.
v1.0 (2026-05). Initial public release. Adult women's kimono, adult men's kimono, simple yukata variant. Fabric widths 33–40, 100–120, 130–160 cm. Default ease 4 cm (women), 6 cm (men). Standard wasai seam allowances. No ohashori tuck handling.
v1.1 (planned). Bug-fix release. Refines the sleeve-width clamp behaviour and tightens the upper-edge collar taper.
v2.0 (planned). Adds juban (under-kimono) and haori (jacket) generators. Adds women's formal ohashori tuck handling. Adds an obi-width helper.
v3.0 (longer-term). Adds children's age-fold preset. Possibly adds hakama (split-skirt trouser). No timeline.
Source code availability
The website's source code is, at the time of writing, private but auditable. The methodology — that is, the formulas above — is fully public. If you find a numerical disagreement between this page and an established source, write and we will share the exact function that produced the output and discuss the source of the disagreement.
The plan is to publish the source code on a permissive open-source license once a small number of wasai-experienced sewers have reviewed the formulas and any obvious errors are fixed. The intermediate position — "open methodology now, open source soon" — is a deliberate choice to avoid putting an unreviewed generator into the world under a label of authority. If you would like to be one of the reviewers, please get in touch via the about page contact channel.
How to send corrections
If you have read this far and disagree with anything on this page, that is the entire point of publishing it. Please send corrections through the contact channel on the about page. The most useful corrections include the dimension you disagree with, the source that supports your figure, and — if you have one — a photograph of a finished garment where the difference matters.
Corrections are reviewed in monthly batches and the methodology page is updated in place with a dated note explaining what changed and why. Sources cited in the correction are added to the source list.