If you’ve ever held two cotton poplins and found one stiff and papery while the other felt almost buttery, you’ve encountered the gap that the Kawabata Evaluation System was built to bridge. The system’s genius isn’t just its sensitive instruments—it’s how those instruments feed linear equations that speak in the language of the hand: koshi, numeri, fukurami. These aren’t vague adjectives; they are mathematical composites of bending, shear, surface friction, and compression, derived through careful regression against expert panels. Understanding that transformation—from a force–deformation curve to a number that says “this fabric is crisp”—is the key to using KES data as a design tool, not a black box.
At its core, the KES primary hand equations are multiple linear regression models. Each primary hand value (HV) is expressed as a sum of weighted mechanical parameters, plus a constant. The general form looks like this:
HV = C₀ + C₁·B + C₂·2HB + C₃·G + C₄·2HG + C₅·2HG5 + C₆·MIU + C₇·MMD + C₈·SMD + C₉·LC + C₁₀·WC + C₁₁·RC + ...
The exact parameters and coefficients depend on the fabric category—a men’s suiting fabric has different sensory priorities than a silk chirimen—but the structure is consistent. Each coefficient distills the contribution of one mechanical property to the hand descriptor, based on the pooled judgments of skilled assessors. When you see a koshi equation heavily weighted on bending rigidity (B) and bending hysteresis (2HB), it’s because stiffness in the hand correlates most strongly with resistance to bending and the energy lost during that bending cycle. The math simply formalizes what fingertips know.
Let’s trace each mechanical curve to the lexicon it feeds. Bending is the most direct. The KES pure-bending tester records a moment–curvature loop; the slope gives bending rigidity B, the width of the loop gives hysteresis 2HB. A fabric that springs back with little hysteresis feels lively, and high B means it resists flexing. Thus koshi—stiffness—draws heavily from B and 2HB, but the coefficients are positive: more rigid bending, higher koshi. This is why a densely woven canvas rates high on koshi, while a limp rayon challis does not.
Shear introduces a different texture. The shear tester measures resistance to in-plane diagonal stretching, producing a force–shear angle curve. The key parameters are shear stiffness G and hysteresis values 2HG (at small angle) and 2HG5 (at larger angle). These capture how much a fabric fights deformation and how much energy it dissipates. In the primary hand equations, shear terms appear strongly in numeri (smoothness) and fukurami (fullness/softness). Numeri is associated with a fluid, drapey hand with little surface friction—so the shear stiffness enters negatively: a low G and low hysteresis allow the fabric to glide and drape smoothly. On the other hand, a certain amount of controlled shear hysteresis can enhance fukurami by giving the fabric a springy, plush resilience—the kind of fullness you feel in a well-finished wool flannel.
Surface properties are measured by two probes mimicking fingertip movement. Friction coefficient MIU and its deviation MMD capture the stick–slip sensation; geometric roughness SMD measures the physical height variation. These terms feed directly into descriptors like shari (crispness) and numeri. Shari is the rustling, granular feel of a raw silk or a dry linen—high SMD and high MMD push shari upward. Numeri, conversely, is correlated with a low, uniform MIU and low SMD: a smooth, almost slippery surface. The equations assign opposite signs to these surface terms depending on the hand being predicted. A fabric can’t be both highly crisp and highly smooth; the math reflects that sensory trade-off.
Compression brings the final dimension. The compression tester records pressure–thickness reduction, yielding linearity LC, compressional energy WC, and resilience RC. These directly map to fukurami—the sense of bulk, springiness, and soft fullness. High LC means the fabric is initially soft but firms up quickly; high WC means it takes more energy to compress; high RC means it bounces back. Together, they create the sensation of stuffing that is both yielding and resilient. A lofty wool melton with high WC and RC will score high on fukurami, while a thin, hard-finished cotton broadcloth will score low. The equations weight these compression parameters positively for fukurami, often in conjunction with moderate bending and shear terms that prevent the fabric from feeling merely limp.
It’s important to see the linear combination not as an arbitrary formula, but as an operationalization of expert consensus. The regression was built by having a panel rank dozens of fabrics on each hand descriptor, then fitting the KES parameters to those ranks. The coefficients are the quantitative expression of a cultural and perceptual calibration—the Japanese hand lexicon. This is why koshi, numeri, and fukurami are not simply translations of “stiff”, “smooth”, and “full”; they encapsulate a particular aesthetic tradition. When we apply these equations to a new fabric, we are in effect asking: how would that tradition judge this material?
For a small label sourcing a deadstock cotton, this means you can read a KES report and know, before the swatch arrives, that the fabric’s high B and moderate G will give it a crisp, structured koshi, but its low SMD might still lend a hint of numeri—so it’s a poplin with a smooth finish, ideal for a shirt collar. Or, conversely, that a high MIU and high 2HG5 spell a granular, energy-absorbing hand that will drape with a dry, almost sculptural quality—perfect for a deconstructed jacket. The equations don’t replace the touch; they give it a shared vocabulary, traceable to physical cause.
This is the translation we need to internalize: every mechanical curve is a silent contributor to a hand value, and every hand value is a pointer back to the material decisions—fiber, yarn twist, weave density, finish—that shaped those curves. The primary hand equations are the bridge, and walking it in both directions turns fabric selection from intuition into insight.
Comments
No comments yet — be the first.