Top > Research > Human Centered Design > Digital Hand Family based on Hand Anthropometry

research

Digital Hand Family based on Hand Anthropometry

Overview

A hand version of the computer manikin, the digital hand, is considered an effective assessment tool for hand-operated products at the design stage. For successful assessment of a product, various sizes and proportions of hands need to be virtualized in a computer. A digital hand family consisting of 9 shapes, covering 95% of Japanese individual differences in hand shapes, was generated from hand measurements to represent hand shapes on the periphery of a distribution.

Boundary Family of Computer Manikins

This is a method to calculate body dimensions of representative forms for computer manikins by Bittner et al. (1986). Body dimensions are correlated with each other. By applying factor analysis, the information carried by the body dimensions is then summarized into 2 uncorrelated factors. By drawing a distribution map of subjects based on the obtained scores of the first 2 factors, a probability ellipsoid is defined in which 95% of the subjects will fall. The center of the distribution is the mean (#9). From the formula of the probability ellipsoid and the formulae of the X-axis (Y=0), the Y-axis (X=0), and Y=±X, we calculate the factor scores of the shapes which are at the intersecting points of the ellipsoid and the 2 factor axes (#1, #5, #3, #7), of the shapes at the intersecting points of the ellipsoid and Y=±X (#2, #4, #6, #8).

Boundary Family of Computer Manikins

Shape Measurement of Hands

By using traditional methods, 25 dimensions were measured for 35 males and 22 females. On the assumption of a link structure, shown in the example drawing to the right, the dimensions to estimate the lengths of phalangeal joint links were measured. Factor analysis was conducted using a total of 39 measurement items.

Shape Measurement of Hands Shape Measurement of Hands Shape Measurement of Hands

Top ∆

The Results of a Factor Analysis

The first factor was related to finger length, and the second factor was related to the joint breadth and hand thickness. On the distribution map: Finger length increases in the positive direction on the X-axis, and the finger thickness increases in the positive direction on the Y-axis.

There was a clear distinction between males and females. If an industrial product is to be designed for separate genders (like fashion gloves), there may be a need to make digital hands for each gender. However, most hand-operated industrial products are unisex in design. The dimensions of the boundary family, shown on the right side of the drawing, are calculated. The mean shape (#9) is at the edge of the distribution for both males and females.

The Results of a Factor Analysis

Digital Hand Family

A commercially available software, Poser (Curious Lab. Inc.), was used to display the calculated dimensions.

Digital Hand Family

References

Top ∆