1. No category

Evaluation of the Multivariate Accommodation Performance of the Grid Method
Kihyo Jung a, Ochae Kwon b, and Heecheon You c,*
a
The Harold and Inge Marcus Department of Industrial and Manufacturing Engineering, The
Pennsylvania State University, University Park, PA 16802, USA
b
Samsung Electronics Co., LTD. 10F, Samsung Electronics Bldg., Seocho 2-dong, Seochogu, Seoul 137-857, South Korea
c
Department of Industrial and Management Engineering, Pohang University of Science and
Technology, Pohang, Kyungbuk 790-784, South Korea
* Corresponding author. Tel.: +82-54-279-2210. Fax: +82-54-279-2870.
E-mail address: hcyou@postech.ac.kr
Evaluation of the Multivariate Accommodation Performance of the Grid Method
Abstract
The present study examined the multivariate accommodation performance (MAP) of the grid
method, a distributed representative human models (RHM) generation method, in the context
of men’s pants sizing system design. Using the 1988 US Army male anthropometric data and
±2.5 cm of fitting tolerance, the grid method selected two key dimensions (waist girth and
crotch height) out of 12 anthropometric dimensions and identified 25 RHMs to accommodate
95% of the population. The average MAP of the RHMs decreased dramatically as the number
of anthropometric dimensions considered increased (99% for single dimension and 14% for
12 dimensions). A standardized regression model was established which explains the effects
of two factors (sum of anthropometric dimension ranges; adjusted R2 between key
dimensions and other anthropometric dimensions) on the MAP of RHMs. This regression
model can be used to prioritize anthropometric dimensions for efficient MAP improvement of
men’s pants design.
Keyword: Multivariate accommodation performance; Grid method; Representative human
model generation; Anthropometry
Introduction
In ergonomic design and evaluation, a small group of human models which represents the
anthropometric variability of the target population is commonly used. Representative human
models (RHMs) are determined by considering the anthropometric characteristics of the
population and a designated accommodation percentage (e.g., 90%) (HFES 300, 2004; Jung
et al., 2009). For example, You et al. (1997) used three RHMs (5th, 50th, and 95th percentiles)
which accommodate 90% of the target population to evaluate an ergonomic design of a bus
operator’s workstation. Use of a small group of RHMs provides an efficient way to develop a
product design which is fitting for the population. However, some studies use a rather large
group of human models (e.g., 6 to 45 manikins) to better reflect the diversity of
anthropometric sizes for ergonomic design and evaluation (Bittner, 2000; Case et al., 2001;
Case et al., 2009; Jung et al., 2009).
Depending on the characteristics of a product being designed, RHMs are determined on
a boundary (Figure 1.a) or a set of grids (Figure 1.b) formed on selected anthropometric
dimensions to accommodate a designated percentage of the population. In designing a onesize product (one size to accommodate people within a designated percentage of the
population) such as a bus operator’s workstation and a helicopter cockpit, RHMs are selected
on an accommodation boundary. For example, Bittner (2000) generated 17 RHMs at the
centroid and the boundary of a hyper-ellipsoid which encompasses 90% of the population for
a workspace design. On the other hand, in designing a multiple-size product (n sizes to fit n
groups of people within a designated percentage of the population) such as gloves and shoes,
RHMs are chosen over a set of grids formed in the distribution of anthropometric dimensions.
For example, Kwon et al. (2009) and Robinette and Annis (1986) constructed grids which
accommodated a designated percentage of the population and then defined representative
cases at the centroids of the grids.
[Figure 1 about here]
Of the various RHM generation methods, the grid method determines RHMs by
following a three-step procedure: selection of key dimensions, formation of representative
grids, and generation of RHMs (Figure 2). First, a small, manageable number of
anthropometric dimensions (e.g., one to five dimensions) are selected as key dimensions by
analyzing the statistical relationships between anthropometric dimensions under
consideration (Gordon and Freill, 1994; Hidson, 1991; Resonblad-Wallin, 1987; Zheng et al.,
2007). Second, representative grids with a fitting tolerance are formed over the distribution of
the key dimensions to accommodate a designated percentage of the population. The value of
fitting tolerance is specified by the product designer by considering various practical aspects
including production economy and material properties (Kwon et al., 2009). Lastly, the values
of the centroid of each representative grid are used for the key dimensions of an RHM and
then regression equations having the key dimensions as regressors are applied to estimate the
other anthropometric dimensions of the RHM (Robinette and Annis, 1986).
[Figure 2 about here]
Research which evaluates the accommodation performance of the grid method in a
comprehensive manner is lacking. Most studies such as Chung et al. (2007) and McCulloch et
al. (1998) using the grid method have considered the multivariate accommodation
performance (MAP) of the grid method only for key dimensions, but not for non-key
anthropometric dimensions. Since some non-key anthropometric dimensions have low
correlations with key dimensions but still affect the fit of product to the population, the MAP
of the grid method needs to be analyzed for all anthropometric dimensions considered in the
design.
The present study examined the MAP of the grid method in a comprehensive manner
and the effects of two factors (sum of anthropometric dimension ranges; adjusted R2 between
key dimensions and other anthropometric dimensions) on MAP. The measure MAP was
quantified by calculating the proportion of the population belonging to grids formed to
accommodate a designated percentage of the population. The two MAP factors were
identified from our empirical observation as possible attributes affecting the MAP of the grid
method. Lastly, standardized multiple regression analysis was conducted to examine the
relative effects of the two factors on MAP.
Methods
Selection of anthropometric dimensions
Referring to Chung et al. (2007) and Hsu and Wang (2005), the present study selected 12
anthropometric dimensions (Table 1) for the design of a men’s pants sizing system. The 1988
US Army anthropometric data of 1774 men (Gordon et al., 1988) was used to generate RHMs.
The anthropometric data was randomly divided into (1) a learning data set of 1,000 cases for
generation of RHMs and (2) a testing data set of 774 cases for cross-evaluation of the
accommodation performance of the generated RHMs.
[Table 1 about here]
Determination of key dimensions
To determine the number of key dimensions for a men’s pants sizing system, maximum
average adjusted R2 was analyzed for different numbers of key dimension candidates as
displayed in Figure 3. The 12 anthropometric dimensions were used to form sets of key
dimension candidates consisting of 1 to 11 anthropometric dimensions. Then, multiple
regression analysis was conducted using each set of key dimension candidates as regressors
and non-key anthropometric dimensions as dependent variables. Finally, for each number of
key dimension candidates, the maximum of average adjusted R2 values were identified. Table
2 illustrates the identification process of the maximum of average adjusted R2 values for
single key dimension candidate cases: for each individual key dimension candidate from AD1
to AD12, adjusted R2 values of regression equations were obtained for the other
anthropometric dimensions and their average adjusted R2 value was calculated, and then the
maximum of the adjusted R2 values was found 0.43 at AD1. Figure 3 shows that the increase
rate of maximum average adjusted R2 becomes significantly small (< 0.1) after two; thus, this
study determined that two anthropometric dimensions would be selected for key dimensions
of a men’s pants sizing system.
[Figure 3 about here]
[Table 2 about here]
Based on the maximum average adjusted R2 analysis results and a survey on key
dimensions used in the garment industry, crotch height (AD5) and waist girth (AD7) were
chosen as key dimensions of a men’s pants sizing system design. The key dimension survey
conducted on the internet in the present study identified that AD5 and AD7 are widely used
for men’s pants sizing in the US garment industry (Banana Republic, 2009; Polo, 2009; Perry
Ellis, 2009).
Calculation of multivariate accommodation performance
The MAP of a particular set of anthropometric dimensions was quantified by calculating the
proportion of anthropometric cases belonging to grids that were formed along generated
RHMs. A fitting tolerance of ±2.5 cm was used as the length of a grid for each
anthropometric dimension by referring to the garment industry practices for the key
dimensions of men’s pants (Banana Republic, 2009; BlueFly, 2009; Polo, 2009; Perry Ellis,
2009). The MAP analysis was conducted for various combinations of anthropometric
dimensions from 1 to 12 using a program coded by Matlab 7.0 (MathWorks, Inc., Natick, MA,
USA). For example, when the number of anthropometric dimensions was 2, MAPs were
calculated for 66 combinations (12C2) of anthropometric dimensions.
Results
Using the grid method, 25 RHMs (Table 3) were generated for the design of a men’s pants
sizing system to accommodate 95% of the population. Square grids with a fitting tolerance of
±2.5 cm were formed which accommodate 95% of the population anthropometric cases
(Figure 4) on the distribution of AD5 and AD7. The centroids of the grids were used for the
key dimension values of the RHMs, and then the regression equations in Table 4, having the
key dimensions as regressors, were applied to determine the values of the other
anthropometric dimension values for the RHMs.
[Table 3 about here]
[Figure 4 about here]
[Table 4 about here]
The MAP analysis (described in the Methods section) showed that the average MAP of
the generated RHMs decreased greatly as the number of anthropometric dimensions
increased as displayed in Figure 5. The univariate accommodation percentage of each
individual anthropometric dimension was higher (96.6% ~ 99.9%) than the target
accommodation percentage (95%); however, the average MAP decreased dramatically as the
number of anthropometric dimensions increased and finally dropped to 14% when all 12
dimensions were considered.
[Figure 5 about here]
During the process of MAP analysis, two factors (sum of anthropometric dimension
ranges, SR; average adjusted R2 between key dimensions and other anthropometric
dimensions, AR) were presumed as those affecting the MAP of the grid method. It was
observed that SR tended to negatively relate to MAP. For example, the MAP of waist-to-knee
length (AD3) (range = 21 cm), waist girth (AD7) (range = 47.5 cm), and ankle girth (AD12)
(range = 9 cm), of which SR = 77.5 cm, was 94%, while that of crotch length (AD6) (range =
38.1 cm), hip girth (AD8) (range = 43.4 cm), and thigh girth (AD9) (range = 32.9 cm), of
which SR = 104.4 cm, was 70% (note that the range data are from Table 1). In contrast, it was
observed that AR tended to positively relate to MAP. For example, the MAP of waist height
(AD1) (adjusted R2 = 0.89), AD3 (adjusted R2 = 0.70), and outside leg length (AD4) (adjusted
R2 = 0.90), of which AR = 0.84, was 95%, while that of waist-ankle length (AD2) (adjusted
R2 = 0.88), AD6 (adjusted R2 = 0.64), and AD9 (adjusted R2 = 0.62), of which AR = 0.71, was
64%.
The statistical significance and relative influence of the two MAP factors were further
examined by multiple regression analysis. A standardized regression equation (F(2, 217) =
95.4, p < 0.001; adjusted R2 = 0.46) was obtained by the stepwise approach (probabilities to
enter and to remove = 0.05 and 0.1):
MAP (%) = (1.03 - 0.616 × SR + 0.284 × AR) × 100
(Eq. 1)
The regression coefficients indicating the directionalities of the factor effects agreed with the
aforementioned observations: SR negatively relates to MAP, while AR positively relates to
MAP. Also, the standardized regression equation showed that SR is more influential to MAP.
Discussion
The present study examined the MAP of the grid method for all the combinations of both key
and non-key anthropometric dimensions. Previous research has evaluated the MAP of the
grid method only for key anthropometric dimensions (Chung et al., 2007; McCulloch et al.,
1998), but not for non-key anthropometric dimensions. Although key dimensions mainly
determine the fitness of a product and an appropriate size of a product for the user (Kwon et
al., 2009), non-key anthropometric dimensions still affect its fitness; thus, an understanding
of the accommodation performance of the grid method is necessary for non-key
anthropometric dimensions.
The MAP analysis results in the present study showed that the grid method can over-
and under-accommodate a designated accommodation percentage of the population
depending on anthropometric dimensions considered. In addition, relatively large decrease
rates (> 10%) or variabilities (> 20%) in accommodation performance were observed in the
mid range of the number of anthropometric dimensions (3 to 8) in which the MAPs for all the
combinations of the 12 anthropometric dimensions were analyzed in the study.
The MAP quantification method used in the present study is applicable to
accommodation performance evaluation of any RHM generation methods. Previous research
has developed various statistical and optimization methods to generate RHMs over the target
population distribution. For example, Eynard et al. (2000) and Laing et al. (1999) used cluster
analysis to classify the target population into representative figure types and then generate
RHMs for each figure type; McCulloch et al. (1998) applied an optimization algorithm under
given constraints (e.g., number of RHMs and fitting tolerance) to generate RHMs. However,
comparison between RHM generation methods has not been made in terms of MAP. An indepth MAP analysis is required in the future for existing RHM generation methods.
The value of fitting tolerance in MAP analysis is often determined by considering
product fitness and production economy (Kwon et al., 2009; McCulloch et al., 1998). A small
value of fitting tolerance can increase the level of fit for a product to the users, but requires a
large number of size categories which negatively affects production economy; the opposite
becomes true for a large value of fitting tolerance.
Different values of fitting tolerance can be used in MAP analysis. The present study
applied ±2.5 cm as a fitting tolerance uniformly to all the anthropometric dimensions for
illustration purposes. However, a value of fitting tolerance for a key or non-key
anthropometric dimension would differ depending on specific design conditions such as
material properties, production economy considerations, wearing purposes, and design
concepts of clothing. Thus, various values of fitting tolerance need to be applied to
anthropometric dimensions when applying the proposed MAP analysis procedure by
considering the design context.
A standardized multiple regression model was established to examine the effects of SR
and AR on MAP. The MAP regression model showed that the less the SR, the higher the
MAP and that the opposite is held for AR. The MAP regression model can be used to analyze
the prioritization of design improvements for better accommodation performance of a product
design. For example, material changes or adjustment mechanisms can be examined with a
higher priority for design parameters that are related to anthropometric dimensions having a
larger SR and/or a smaller AR. To obtain a more comprehensive, optimal solution for design
improvement in industry, cost factors would be taken into account in addition to MAP
sensitivity results.
Acknowledgments
This research was supported by Basic Science Research Program through the National
Research Foundation of Korea (NRF) funded by the Ministry of Education, Science, and
Technology (2010-0012291).
References
Banana Republic (2009). Men’s pant sizing system. Retrieved July 10, 2009 from
http://www.gap.com/browse/category.do?cid=11889.
Bittner, A. C. (2000). A-CADRE: Advanced family of manikins for workstation design. In
Proceedings of the IEA 2000/HFES 2000 Congress. San Diego. CA, 774-777.
BlueFly (2009). Men’s pant sizing system. Retrieved July 10, 2009 from
http://www.bluefly.com/Theory-wheat-khaki-wool-MarloWythetrousers/cat60015/304852501/detail.fly.
Case, K., Marshall, R., Hogberg, D., Summerskill, S., Gyi, D., and Sims, R. (2009).
HADRIAN: Fitting Trials by Digital Human Modelling. Lecture Notes In Computer
Science, 673 – 680.
Case, K., Porter, M., Gyi, D., Marshall, R., and Oliver, R. (2001). Virtual fitting trials in
design for all. Journal of Materials Processing Technology, 117, 255 - 261.
Chung, M., Lin, H., and Wang, M. (2007). The development of sizing systems for Taiwanese
elementary- and high-school students. International Journal of Industrial Ergonomics,
37(8), 707-716.
Eynard, E., Fubini, E., Masali, M., Cerrone, M., and Tarzia, A. (2000). Generation of virtual
man models representative of different body proportion and application to ergonomic
design of vehicles. In Proceedings of the IEA2000/HFES2000 Congress. San Diego. CA,
489-492.
Gordon, C. C., Bradtmiller, B., Churchill, T., Clauser, C., McConville, J., Tebbetts, I., and
Walker, R. (1988). 1988 Anthropometric Survey of US Army Personnel: Methods and
Summary Statistics (Technical Report NATICK/TR-89/044), US Army Natick Research
Center: Natick, MA.
Gordon, C. C., and Friedl, K. E. (1994). Anthropometry in the US armed forces. In Ulijaszek,
S. J. (ed.). Anthropometry: The Individual and the Population. Cambridge University
Press: Cambridge UK.
HFES 300 (2003). Guidelines for Using Anthropometric Data in Product Design. Santa
Monica, California: Human Factors and Ergonomics Society.
Hidson, D. (1991). Development of a Standard Anthropometric Dimension Set for Use in
Computer-Aided Glove Design (DREO Technical Note 91-22). Defense Research
Establishment OTTAWA.
Hsu, C., and Wang, M. (2005). Using decision tree-based data mining to establish a sizing
system for the manufacture of garments. International Journal of Advanced
Manufacturing Technology, 26. 669-674.
Jung, K., Kwon, O., and You, H. (2009). Development of a digital human model generation
method for ergonomic design in virtual environment. International Journal of Industrial
Ergonomics. 39 (5), 744-748.
Jung, K., You, H., and Kwon, O. (2008). A multivariate evaluation method for representative
human model generation methods: Application to grid method. In Proceedings of the
Human Factors and Ergonomics Society 52nd Annual Meeting. New York: The Human
Factors and Ergonomics Society, 1665-1669.
Kwon, O., Jung, K., You, H., and Kim, H. (2009). Determination of key dimensions for a
glove sizing system by analyzing the relationships between hand dimensions. Applied
Ergonomics, 40 (4), 762-766.
Laing, R. M., Holland, E. J., Wilson, C. A., and Niven, B. E. (1999). Development of sizing
systems for protective clothing for the adult male. Ergonomics, 42(10), 1249-1257.
McCulloch, C. E., Paal, B., and Ashdown, S. P. (1998). An optimization approach to apparel
sizing. Journal of the Operational Research Society, 49(5), 492-299.
Polo (2009). Men’s pant sizing system. Retrieved July 10, 2009 from
http://www.ralphlauren.com/product/index.jsp?productId=3678183&cp=1760781.176081
1&ab=ln_men_cs1_pants&parentPage=family.
PerryEllis (2009). Men’s pant sizing system. Retrieved July 10, 2009 from
http://www.perryellis.com/Pants/Micro-TwillPant/invt/49mb8308ps&bklist=icat,4,shop,pemen,pedmbott.
Robinette, K. M., and Annis, J. F. (1986). A Nine-Size System for Chemical Defense Gloves.
Technical Report (AAMRL-TR-86-029) (ADA173 193). Harry G. Armstrong Aerospace
Medical Research Laboratory, Wright-Patterson Air Force Base, OH.
Rosenblad-Wallin, E. (1987). An anthropometric study as the basis for sizing anatomically
designed mittens. Applied Ergonomics, 18(4), 329-333.
You, H., Bucciaglia, J., Lowe, B. D., Gilmore, B. J., and Freivalds, A. (1997). An ergonomic
design process for a US transit bus operator workstation. Heavy Vehicle Systems, A
Series of the International Journal of Vehicle Design, 4(2-4), 91-107.
Zheng, R., Yu, W., and Fan, J. (2007). Development of a new Chinese bra sizing system
based on breast anthropometric measurements. International Journal of Industrial
Ergonomics, 37(8), 697-705.
List of Figures
Figure 1. Determination of representative cases (small dots: population cases; large dots:
representative cases)
Figure 2. Representative human model (RHM) generation process of the grid method (AD:
anthropometric dimension; K: key dimension)
Figure 3. The trend of maximum adjusted R2 by the number of key dimensions
Figure 4. Formation of representative grids (fitting tolerance = ±2.5 cm) accommodating 95%
of the 1988 US Army male population (small dots: population cases; large dots:
representative cases)
Figure 5. Average and range (min and max) of accommodation percentage by the number of
anthropometric dimensions
List of Tables
Table 1. Anthropometric dimensions selected for design of men’s pants sizing system
Table 2. Analysis of average adjusted R2 for single key dimension candidates
Table 3. Representative human models generated by the grid method to accommodate 95% of
the 1988 US Army male population
Table 4. Regression equations using crotch height (AD5) and waist girth (AD7) as regressors
(a) Boundary method
(b) Grid method
Figure 1. Determination of representative cases (small dots: population cases; large dots:
representative cases) (illustrated)
Step 1: Selection of
key dimensions
Step 3: Generation of
representative human models
Step 2: Formation of
representative grids
Frequency analysis
AD1 = f1 ( K1 , K 2 )
AD1
AD2
.
.
.
ADn
Reducing
variables
K1
K2
RHM1
RHM2
RHM3
.
.
.
K2
AD2 = f 2 ( K1 , K 2 )
.
.
.
ADn = f n ( K1 , K 2 )
Tolerance
K1
Correlation analysis
Factor analysis
Frequency analysis
Regression analysis
Figure 2. Representative human model (RHM) generation process of the grid method (AD:
anthropometric dimension; K: key dimension)
2 with other dimensions
Maximum adjusted
AverageR
of adjusted
R2
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
1
2
3
4
5
6
7
8
9
10
11
Number of key dimensions
Figure 3. The trend of maximum adjusted R2 by the number of key dimensions
Figure 4. Formation of representative grids (fitting tolerance = ±2.5 cm) accommodating 95%
of the 1988 US Army male population (small dots:
representative cases) (illustrated)
population cases; large dots:
100
Accommodation (%)
99
80
93
82
69
60
57
40
46
37
30
20
24
20
16
14
11
12
0
1
2
3
4
5
6
7
8
9
10
Number of anthropometric dimensions
Figure 5. Average and range (min and max) of accommodation percentage by the number of
anthropometric dimensions
Table 1. Anthropometric dimensions selected for design of men’s pants sizing system
Dimensional
type
Length
and height
Girth
a
Code
Selected dimensions
AD1
AD2
AD3
AD4
AD5
AD6
AD7
AD8
AD9
AD10
AD11
AD12
Waist height
Waist-to-ankle lengtha
Waist-to-knee lengtha
Outside leg length
Crotch height
Crotch length
Waist girth
Hip girth
Thigh girth
Knee girth
Calf girth
Ankle girth
Hsu and
Chung et al.
Wang (2005)
(2007)
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
Descriptive statisticsb
Mean
SD
Range
112.7
5.2
43.1
106.0
5.0
40.3
56.8
2.7
21.0
108.2
5.1
41.0
83.7
4.6
39.2
76.7
5.6
38.1
84.0
7.4
47.5
98.4
6.2
43.4
59.7
4.9
32.9
38.6
2.2
14.7
37.8
2.5
16.6
22.2
1.3
9.0
Waist-to-ankle length and waist-to-knee length were estimated by subtracting ankle height
and knee height from waist height, respectively, because of their unavailability in the 1988
US Army anthropometric data.
b
Source: The 1988 US Army anthropometric data of males (n = 1774) (Gordon et al., 1988)
Table 2. Analysis of average adjusted R2 for single key dimension candidates (illustrated)
Key dimension
candidate
AD1
AD2
M
AD12
Anthropometric
dimensions
AD2
AD3
AD4
AD5
AD6
AD7
AD8
AD9
AD10
AD11
AD12
AD1
AD3
Adjusted R2
M
M
AD12
0.13
M
M
AD1
AD2
0.15
0.13
M
M
AD11
0.64
0.99
0.89
0.88
0.83
0.20
0.10
0.18
0.12
0.23
0.12
0.15
0.99
0.89
Average
adjusted R2
0.43
0.42
M
0.28
Table 3. Representative human models generated by the grid method to accommodate 95% of the 1988 US Army male populationa
Non-key anthropometric dimensionb
Key dimensions
No
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Crotch
height
(AD5)
73.8 (2)
73.8 (2)
73.8 (2)
78.8 (15)
78.8 (15)
78.8 (15)
78.8 (15)
78.8 (15)
78.8 (15)
83.8 (51)
83.8 (51)
83.8 (51)
83.8 (51)
83.8 (51)
83.8 (51)
83.8 (51)
88.8 (87)
88.8 (87)
88.8 (87)
88.8 (87)
88.8 (87)
88.8 (87)
93.8 (99)
93.8 (99)
93.8 (99)
Waist girth
(AD7)
74.1 (9)
79.1 (25)
84.1 (50)
69.1 (2)
74.1 (9)
79.1 (25)
84.1 (50)
89.1 (75)
94.1 (91)
69.1 (2)
74.1 (9)
79.1 (25)
84.1 (50)
89.1 (75)
94.1 (91)
99.1 (98)
74.1 (9)
79.1 (25)
84.1 (50)
89.1 (75)
94.1 (91)
99.1 (98)
79.1 (25)
84.1 (50)
89.1 (75)
Waist
height
(AD1)
101.0 (1)
101.9 (2)
102.8 (3)
105.2 (7)
106.0 (10)
106.9 (13)
107.8 (17)
108.7 (22)
109.6 (27)
110.2 (31)
111.1 (38)
112.0 (44)
112.9 (51)
113.7 (58)
114.6 (64)
115.5 (70)
116.1 (74)
117.0 (80)
117.9 (84)
118.8 (88)
119.7 (91)
120.5 (93)
122.0 (96)
122.9 (98)
123.8 (98)
Waist-toankle
length
(AD2)
94.8 (1)
95.6 (2)
96.4 (3)
98.9 (8)
99.7 (10)
100.5 (14)
101.3 (17)
102.1 (22)
102.9 (27)
103.7 (32)
104.5 (38)
105.3 (45)
106.1 (51)
106.9 (57)
107.8 (64)
108.6 (70)
109.4 (75)
110.2 (80)
111.0 (84)
111.8 (88)
112.6 (91)
113.4 (93)
115.0 (96)
115.8 (98)
116.6 (98)
Waist-toknee
length
(AD3)
51.3 (2)
51.8 (3)
52.3 (5)
53.2 (9)
53.7 (12)
54.1 (16)
54.6 (21)
55.0 (26)
55.5 (31)
55.5 (32)
56.0 (38)
56.4 (44)
56.9 (51)
57.4 (58)
57.8 (64)
58.3 (70)
58.3 (70)
58.8 (76)
59.2 (81)
59.7 (85)
60.1 (89)
60.6 (92)
61.1 (94)
61.5 (96)
62.0 (97)
Outside
leg length
(AD4)
96.8 (1)
97.6 (2)
98.4 (3)
100.9 (8)
101.7 (10)
102.6 (13)
103.4 (17)
104.2 (22)
105.1 (27)
105.8 (32)
106.7 (38)
107.5 (45)
108.4 (51)
109.2 (58)
110.0 (64)
110.9 (70)
111.6 (75)
112.5 (80)
113.3 (84)
114.2 (88)
115.0 (91)
115.8 (93)
117.4 (96)
118.3 (98)
119.1 (98)
Crotch
length
(AD6)
69.9 (11)
72.9 (25)
75.8 (44)
67.4 (5)
70.3 (13)
73.3 (27)
76.3 (47)
79.2 (68)
82.2 (84)
67.8 (6)
70.8 (14)
73.7 (30)
76.7 (50)
79.7 (71)
82.6 (86)
85.6 (95)
71.2 (16)
74.2 (33)
77.2 (53)
80.1 (73)
83.1 (88)
86.1 (95)
74.6 (36)
77.6 (57)
80.6 (76)
Hip girth
(AD8)
89.5 (8)
93.0 (19)
96.4 (38)
87.0 (3)
90.5 (10)
94.0 (24)
97.4 (44)
100.9 (66)
104.4 (83)
88.0 (5)
91.5 (13)
95.0 (29)
98.4 (50)
101.9 (72)
105.4 (87)
108.8 (95)
92.5 (17)
96.0 (35)
99.4 (57)
102.9 (77)
106.4 (90)
109.8 (97)
97.0 (41)
100.5 (63)
103.9 (81)
a
The values are in cm and corresponding percentile values are presented in parentheses.
b
Non-key dimensions were estimated by regression equations in Table 4.
Thigh
girth
(AD9)
53.5 (11)
56.1 (24)
58.7 (42)
51.4 (5)
54.0 (13)
56.6 (27)
59.2 (46)
61.8 (67)
64.4 (83)
51.9 (6)
54.5 (15)
57.1 (30)
59.7 (50)
62.3 (70)
64.9 (86)
67.5 (94)
55.0 (17)
57.6 (34)
60.2 (54)
62.8 (74)
65.4 (88)
68.0 (95)
58.1 (38)
60.7 (58)
63.3 (77)
Knee girth
(AD10)
35.5 (8)
36.5 (17)
37.5 (31)
35.0 (5)
36.0 (12)
37.1 (24)
38.1 (40)
39.1 (59)
40.2 (75)
35.6 (8)
36.6 (18)
37.6 (33)
38.7 (50)
39.7 (68)
40.7 (83)
41.8 (92)
37.2 (25)
38.2 (42)
39.2 (61)
40.3 (77)
41.3 (88)
42.3 (95)
38.8 (53)
39.8 (70)
40.8 (84)
Calf girth
(AD11)
34.9 (13)
36.0 (24)
37.1 (39)
34.2 (8)
35.3 (16)
36.4 (29)
37.5 (45)
38.6 (62)
39.7 (77)
34.6 (10)
35.6 (20)
36.7 (34)
37.8 (50)
38.9 (67)
40.0 (81)
41.1 (90)
36.0 (24)
37.1 (39)
38.2 (56)
39.3 (72)
40.4 (84)
41.5 (92)
37.5 (45)
38.5 (61)
39.6 (76)
Ankle
girth
(AD12)
20.7 (14)
21.2 (22)
21.6 (34)
20.6 (11)
21.0 (19)
21.5 (29)
21.9 (42)
22.3 (55)
22.8 (68)
20.9 (16)
21.3 (25)
21.7 (37)
22.2 (50)
22.6 (64)
23.1 (75)
23.5 (85)
21.6 (33)
22.0 (46)
22.5 (59)
22.9 (71)
23.3 (82)
23.8 (89)
22.3 (54)
22.7 (67)
23.2 (78)
Table 4. Regression equations using crotch height (AD5) and waist girth (AD7) as regressors
Anthropometric
dimensions
Waist height
Waist-to-ankle length
Waist-to-knee length
Outside leg length
Crotch length
Hip girth
Thigh girth
Knee girth
Calf girth
Ankle girth
Code
AD1
AD2
AD3
AD4
AD6
AD8
AD9
AD10
AD11
AD12
Regression equation
AD1 = 13.49 + 1.01°AD5 + 0.18°AD7
AD2 = 11.24 + 0.97°AD5 + 0.16°AD7
AD3 = 10.33 + 0.46°AD5 + 0.09°AD7
AD4 = 11.07 + 0.99°AD5 + 0.17°AD7
AD6 = 19.53 + 0.09°AD5 + 0.59°AD7
AD8 = 23.25 + 0.20°AD5 + 0.69°AD7
AD9 = 7.76 + 0.10°AD5 + 0.52°AD7
AD10 = 11.77 + 0.11°AD5 + 0.21°AD7
AD1 1= 13.47 + 0.07°AD5 + 0.22°AD7
AD12 = 10.04 + 0.06°AD5 + 0.09°AD7
Adjusted R2
0.89
0.88
0.70
0.90
0.64
0.72
0.62
0.55
0.43
0.30