IJEV5I3

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/271520706
Analysis of Hand Anthropometric Dimensions of Male Industrial Workers of

Haryana State
Article · August 2011
CITATIONS READS
15 1,707
1 author:
Arunesh Chandra
Krishna Institute Of Engineering And Technology
36 PUBLICATIONS 249 CITATIONS
SEE PROFILE
All content following this page was uploaded by Arunesh Chandra on 30 January 2015.
The user has requested enhancement of the downloaded file.

INTERNATIONAL JOURNAL OF
ENGINEERING (IJE)
VOLUME 5, ISSUE 3, 2011
EDITED BY
DR. NABEEL TAHIR
ISSN (Online): 1985-2312

International Journal of Engineering is published both in traditional paper form and in Internet.
This journal is published at the website http://www.cscjournals.org, maintained by Computer
Science Journals (CSC Journals), Malaysia.
IJE Journal is a part of CSC Publishers

Computer Science Journals
http://www.cscjournals.org
INTERNATIONAL JOURNAL OF ENGINEERING (IJE)
Book: Volume 5, Issue 3, August 2011

Publishing Date: 31-08-2011
ISSN (Online): 1985-2312
This work is subjected to copyright. All rights are reserved whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting,
re-use of illusions, recitation, broadcasting, reproduction on microfilms or in any
other way, and storage in data banks. Duplication of this publication of parts
thereof is permitted only under the provision of the copyright law 1965, in its
current version, and permission of use must always be obtained from CSC
Publishers.
IJE Journal is a part of CSC Publishers

http://www.cscjournals.org
© IJE Journal
Published in Malaysia
Typesetting: Camera-ready by author, data conversation by CSC Publishing Services – CSC Journals,
Malaysia
CSC Publishers, 2011

EDITORIAL PREFACE
This is the third issue of volume five of International Journal of Engineering (IJE). The Journal is
published bi-monthly, with papers being peer reviewed to high international standards. The
International Journal of Engineering is not limited to a specific aspect of engineering but it is
devoted to the publication of high quality papers on all division of engineering in general. IJE
intends to disseminate knowledge in the various disciplines of the engineering field from
theoretical, practical and analytical research to physical implications and theoretical or
quantitative discussion intended for academic and industrial progress. In order to position IJE as
one of the good journal on engineering sciences, a group of highly valuable scholars are serving
on the editorial board. The International Editorial Board ensures that significant developments in
engineering from around the world are reflected in the Journal. Some important topics covers by
journal are nuclear engineering, mechanical engineering, computer engineering, electrical
engineering, civil & structural engineering etc.
The initial efforts helped to shape the editorial policy and to sharpen the focus of the journal.
Starting with volume 5, 2011, IJE appears in more focused issues. Besides normal publications,
IJE intend to organized special issues on more focused topics. Each special issue will have a
designated editor (editors) – either member of the editorial board or another recognized specialist
in the respective field.
The coverage of the journal includes all new theoretical and experimental findings in the fields of
engineering which enhance the knowledge of scientist, industrials, researchers and all those
persons who are coupled with engineering field. IJE objective is to publish articles that are not
only technically proficient but also contains information and ideas of fresh interest for International
readership. IJE aims to handle submissions courteously and promptly. IJE objectives are to
promote and extend the use of all methods in the principal disciplines of Engineering.
IJE editors understand that how much it is important for authors and researchers to have their
work published with a minimum delay after submission of their papers. They also strongly believe
that the direct communication between the editors and authors are important for the welfare,
quality and wellbeing of the Journal and its readers. Therefore, all activities from paper
submission to paper publication are controlled through electronic systems that include electronic
submission, editorial panel and review system that ensures rapid decision with least delays in the
publication processes.
To build its international reputation, we are disseminating the publication information through
Google Books, Google Scholar, Directory of Open Access Journals (DOAJ), Open J Gate,
ScientificCommons, Docstoc and many more. Our International Editors are working on
establishing ISI listing and a good impact factor for IJE. We would like to remind you that the
success of our journal depends directly on the number of quality articles submitted for review.
Accordingly, we would like to request your participation by submitting quality manuscripts for
review and encouraging your colleagues to submit quality manuscripts for review. One of the
great benefits we can provide to our prospective authors is the mentoring nature of our review
process. IJE provides authors with high quality, helpful reviews that are shaped to assist authors
in improving their manuscripts.
Editorial Board Members

International Journal of Engineering (IJE)
EDITORIAL BOARD
Editor-in-Chief (EiC)
Dr. Kouroush Jenab

Ryerson University (Canada)
ASSOCIATE EDITORS (AEiCs)
Professor. Ernest Baafi

University of Wollongong
Australia
Dr. Tarek M. Sobh

University of Bridgeport
United States of America
Professor. Ziad Saghir

Ryerson University
Canada
Professor. Ridha Gharbi

Kuwait University
Kuwait
Professor. Mojtaba Azhari

Isfahan University of Technology
Iran
Dr. Cheng-Xian (Charlie) Lin

University of Tennessee
EDITORIAL BOARD MEMBERS (EBMs)
Dr. Dhanapal Durai Dominic P

Universiti Teknologi Petronas
Malaysia
Professor. Jing Zhang

University of Alaska Fairbanks
Dr. Tao Chen

Nanyang Technological University
Singapore
Dr. Oscar Hui
University of Hong Kong
Hong Kong
Professor. Sasikumaran Sreedharan

King Khalid University
Saudi Arabia
Assistant Professor. Javad Nematian

University of Tabriz Iran
Dr. Bonny Banerjee

Senior Scientist at Audigence
AssociateProfessor. Khalifa Saif Al-Jabri

Sultan Qaboos University
Oman
Dr. Alireza Bahadori

Curtin University
Australia
Dr Guoxiang Liu
University of North Dakota
Dr Rosli
Universiti Tun Hussein Onn
Malaysia
Professor Dr. Pukhraj Vaya

Amrita Vishwa Vidyapeetham
India
TABLE OF CONTENTS
Volume 5, Issue 3, August 2011
Pages
242 - 256 Analysis of Hand Anthropometric Dimensions of Male Industrial Workers of Haryana State
Arunesh Chandra, Pankaj Chandna, Surinder Deswal
257 - 267 Averaging Method for PWM DC-DC Convertors operating in Discontinuous Conduction
Mode With Feedback
Mohammed S. Al-Numay, N.M. Adamali Shah
268 - 276 Semantic Web Mining of Un-structured Data: Challenges & Opportunities
Manoj Manuja, Deepak Garg
International Journal of Engineering (IJE), Volume (5), Issue (3) : 2011

Arunesh Chandra, Pankaj Chandna & Surinder Deswal
Analysis of Hand Anthropometric Dimensions of Male Industrial

Workers of Haryana State
Arunesh Chandra [email protected]

Faculty of Engineering/Mechanical
Engineering Department/
Krishna Institute of Engineering & Technology
Ghaziabad, UttarPradesh, 201206, India
Pankaj Chandna [email protected]

Faculty of Engineering/Mechanical
National Institute of Technology
Kurushetra, Haryana, 136119, India
Surinder Deswal [email protected]

Faculty of Engineering/Civil
National Institute of Technology
Kurushetra, Haryana, 136119, India
Abstract
The purpose of this paper is to analyse the thirty-seven hand anthropometric characteristics of
the industrial worker of the Haryana state. A survey of convenience sample of eight hundred and
seventy eight male industrial workers was conducted in the year 2009. Paper contains data from
all the four divisions of Haryana state of India and from the five age groups. Minimum, maximum,
mean, standard deviation, skewness, coefficient of variation, 5th, 50th, and 95th percentile for each
hand anthropometric dimension were calculated for the entire state. The normality assumption
was evaluated for each hand dimension, separately. It was found that in most hand dimensions
there were differences between five age groups. Additionally, the statistical analysis was carried
out to correlate various hand dimensions and to obtain prediction equation between different
variables. It has been found that most of the hand dimensions are correlated significantly with
each other. The data gathered may be used for the design of hand tools, gloves, machine access
spaces and hand-held devices and for selection of hand tools for use by Industrial worker
working in the Haryana state of India.
Keywords: Hand Anthropometric Measurements, Industrial Worker, Hand Tools, Prediction

Equation
1. INTRODUCTION
The economic growth and technological improvements have lead to greater demand and
development of machines and devices used in industrial settings. With these dramatic changes
there has also been greater interaction between man and machines. Anthropometric data are one
of essential factors in designing machines and devices as described by [1 & 2]. Incorporating
such information would yield more effective designs, ones that are more user friendly, safer, and
enable higher performance and productivity. According to [3 & 4] the lack of properly designed
machines and equipments may lead to lower work performance and higher incidence to work
related injuries [5] have discussed that for years, anthropometry has been used in national sizing
surveys as an indicator of health status. Anthropometric measurement of human limbs plays an
important role in design of workplace, clothes, hand tools, manual tasks or access spaces for the
hand and many products for human use.
International Journal of Engineering (IJE), Volume (5) : Issue (3) : 2011 242
Many studies have been conducted in the past to study the hand anthropometry. The depth and
breadth of each segment of the hand were measured at points that were spaced at equal
distance between the joints of the hand by [6]. Data on the mean length of the proximal and
middle phalangeal segments for the fingers was published by [7]. Also [8] described that the
interaction of handle size and shape with the kinematics and anthropometry of the hand have a
great effect on hand posture and grip strength. Anthropometric survey measuring 18 dimensions
of the right hand female workers living in Western Nigeria was conducted by [9] and the means of
the collected data were compared with those females from USA, UK and Hongkong. Grip tasks
for six subjects were studied using the hand measurement system by [10] the result showed that
the flexion angle for the five fingers decreased with increasing grip span. [11 & 12] have stressed
the importance of interplay of hand anthropometry and handle size or shape in influencing hand
posture, grip span or grip strength. [13] estimated internal biomechanical loads of the hand from
external loads and finger lengths that were themselves estimated from measured hand length
and breadth; and found that hand anthropometric measurements, especially palm width, are
better predictors of hand strength than stature and body weight. The effects on hand grip forces
by relatively small changes in hand or handle size have also been demonstrated by [14] for
torquing on cylinders [15] for gripping cylinders and [16] for gripping and squeezing on parallel
handles of a standard handgrip dynamometer. Hence, measurement of small difference in hand
size is important in understanding gripping forces. An important implication of the above
discussion is that the anthropometry of the hand must be known for any target population for
whom hand tools and other manual devices are to be designed. [17] stated that today, there is a
growing demand among professional hand tool users to have ergonomically designed product.
Further [18, 19 & 20] have discussed that poor ergonomic hand tools design is a well known
factor contributing to biomechanical stresses and increasing the risk of cumulative trauma and
carpal tunnel syndrome disorders of users. According to [21]) hand anthropometry is useful for
determining various aspects of industrial machineries so as to design the equipment and
machines for better efficiency and more human comfort. [22] discussed the potentially harmful
effects of ignoring anthropometric differences between populations may be manifested when a
developing nation, for example, imports equipment from a developed nation since the latter tends
to design their equipment based on the anthropometric data of their own population. Reliable data
on the association between hand injuries or disorders and hand anthropometry are almost absent
in the developing countries. According to [23 & 24] the continued reliance on muscular power in
tool use, in developing countries, and the widespread use of hand tools that do not fit the hands
properly results in problems of health, safety and task performance. Further data on relevant
anthropometric dimensions of the populations of the importing countries for equipment design
may help alleviate the problems. Only a limited work has been reported in connection of hand
anthropometry data for the populations of developing countries by [25, 26, 27, 28, 29 & 30].
Keeping the above-mentioned factors in consideration, the present analysis is an attempt to study
the impact of collected hand anthropometric data of male industrial worker of Haryana state. As
Haryana state of India has total geographical area of 44212 sq. meters. As per Census data 2001
male population of state is 11364000 with about total 498656 (5%) of male population working in
almost about 72643 registered industrial units with output @ 6430 Crores with major SME (small
manufacturing enterprises) clusters and SEZ (small economic zone) in the Haryana State of
India. These movements and others provide incentives for foreign suppliers and investors to open
factories and service sectors in Haryana state of India. Many of the industries being developed,
therefore, would depend heavily on tools and equipment imported from IC (Industralized
Countries) with the negative consequences as described above, if no attempt will be made to
match equipment design with human characteristics. The present study thus represents an effort
for analysing hand anthropometry data of male industrial worker. The data from this study will
also help to understand the anatomical relationships among the various segments of the hand
within the Haryana Industrial worker population.
2. METHODS
2.1 Subjects and Apparatus
Sets of thirty-seven hand dimensions were measured for each industrial worker. Selection of
these dimensions were made on the basis of their relevance to the design of industrial tools,
machine guarding and other manual equipments, and also because they have been measured in
previous research studies in different populations. The figures of the hand dimensions are
provided in figures 1(a) and 1(b). A total of 878-convenience sample of participants were
measured from thirty-eight small and medium scale industries located in different divisions of the
state. The range included companies from the automobile, tools and instruments, railway
workshop, agricultural and metal sectors, among other, mainly located in the four different
divisions (Ambala, Rohtak, Gurgaon and Hisar) of the Haryana state of the India. Subjects were
selected according to their availability and willingness to participate without payment or any other
kind of reward they were informed with the objectives of the study, anthropometric dimensions,
clothing requirements, measurements procedures and freedom to withdraw. Age of the subjects
varied between 18 and 62 years old with an average age of 37.91 years, whereas average
stature height and body weight of the subjects was found out to be 1653.23 mm and 65.14 kg
respectively. The sample comprised essentially individuals from industry. Underlying the choice of
subjects from industry is the fact that this account for approximately 5% of active adult male
population of Haryana state (Census, 2001). The methods of hand anthropometric measurements
were same as stated by [31 & 32]. Regular measurement tools are used such as Hardenpen
anthropometer for stature measurement and arm length measurement, small anthropometer for
elbow length measurement, digital vernier caliper for length, breadth and depth measurement of
hand, measuring tape for circumferential measurements, a wooden cone designed locally and
specially to measure internal grip diameter, inner caliper for measurement of grip span and the
body weight was measured by portable weighing digital scale. Table 1 describes the age
distribution of the sample of the subjects measured.
FIGURE 1(A): Selected right hand Anthropometric dimensions of Male Industrial worker Defined In Table 3
FIGURE 1(B): Selected right hand Anthropometric dimensions of Male Industrial worker Defined In Table 3
Male Industrial Worker

Age Group (Years)
Number Percentage
18 – 25 133 15.15
26 – 35 253 28.82
36 – 45 221 25.17
46 – 55 218 24.83
56-Above 53 6.04
TABLE 1: Age distribution of Subjects
3. RESULTS
According to [33 & 34] there are many factors in human measurements that intervene as sources
of error and results can be systematically different in spite of the measures being highly trained.
In anthropometric research the measurer cannot perceive the anomalous measures, as the norm
has a very wide range and the size differences among the subjects of a sample are much higher
than the accuracy of experimental devices, sometimes a factor of 10 or higher. Thus the data
collected was further analyzed using SPSS statistical package (version 16.0) for normality
distribution of each hand dimension, using the Kolmogorov-Smirnov and using the Shapiro-Wilk
test at the 5% level of significance, the results of the tests are shown in table 2. Outputs are also
obtained from box-plots generated from the explore command and the extreme outliers that is
1.77% of the collected readings are rejected for further analysis as they are not following the
normal distribution curve as these may be systematic or bias errors which are possible which may
not be clearly noticeable and occasionally these may be systematic errors in the measurement
processes which could have a significant effect on both mean values of experimental variable and
their standard deviation could cause mistaken conclusions over considered population.
Kolmogorov Smirnov Shapiro-Wilk

S.No. Measured Parameter
Statistic Significance Statistic Significance
1 Age 0.072 0.000 0.975 0.000
2 Stature height 0.069 0.000 0.971 0.000
3 Weight 0.052 0.004 0.993 0.030
4 Finger tip to root digit 5 0.041 0.056 0.994 0.042
5 First joint to root digit 5 0.039 0.092 0.994 0.063
6 Second joint to root digit 5 0.049 0.009 0.991 0.008
7 Finger tip to root digit 3 0.063 0.000 0.982 0.000
8 First joint to root digit 3 0.039 0.083 0.991 0.004
9 Second joint to root digit 3 0.048 0.011 0.991 0.006
10 Breadth at tip digit 5 0.035 0.200 0.992 0.013
11 Breadth at first joint digit 5 0.044 0.032 0.993 0.035
12 Breadth at second joint digit 5 0.045 0.025 0.991 0.005
13 Breadth at tip digit 3 0.032 0.200 0.995 0.124
14 Breadth at first joint digit 3 0.050 0.007 0.990 0.003
15 Breadth at second joint digit 3 0.045 0.025 0.991 0.005
16 Depth at tip digit 5 0.055 0.002 0.987 0.000
17 Depth at first joint digit 5 0.042 0.044 0.994 0.075
18 Depth at second joint digit 5 0.070 0.000 0.977 0.000
19 Depth at tip digit 3 0.051 0.006 0.993 0.026
20 Depth at first joint digit 3 0.044 0.031 0.989 0.001
21 Depth at second joint digit 3 0.044 0.032 0.990 0.002
22 Grip span 0.051 0.006 0.994 0.066
23 Max. breadth of the hand 0.092 0.000 0.978 0.000
24 Breadth of the knuckles 0.079 0.000 0.984 0.000
25 Hand length 0.084 0.000 0.986 0.000
26 Palm length 0.077 0.000 0.989 0.002
27 Depth of the knuckles 0.136 0.000 0.963 0.000
28 Max. depth of the hand 0.070 0.000 0.989 0.002
29 Fist length 0.105 0.000 0.987 0.000
30 First phalanx digit 3 length 0.105 0.000 0.967 0.000
31 Fist circumference 0.064 0.000 0.991 0.007
32 Hand circumference 0.072 0.000 0.988 0.000
33 Max. hand circumference 0.057 0.001 0.990 0.003
34 Index finger circumference 0.127 0.000 0.972 0.000
35 Wrist circumference 0.087 0.000 0.990 0.003
36 Arm length 0.077 0.000 0.982 0.000
37 Elbow length 0.050 0.006 0.986 0.000
38 Elbow flexed 0.070 0.000 0.989 0.001
39 Max. internal grip diameter 0.175 0.000 0.934 0.000
40 Middle finger palm grip diameter 0.196 0.000 0.913 0.000
TABLE 2: Comparison of the empirical distribution of the sample vs. the theoretical (Normal) distribution for
Male Industrial Worker
With consideration of normal distribution table 3 provides the minimum, maximum, mean,
standard deviation, coefficient of variation, skewness of each hand dimension and the values of
th th th
each hand dimension at the 5 , 50 , and 95 percentile.
S. Percentile
Hand Skew
N Min. Max. Mean SD CV th
dimensions ness 5 50th 95th
o
Finger tip to root
1 49.79 68.10 59.13 3.39 5.73 -0.117 52.97 59.95 66.89
digit 5
First joint to root
2 27.31 41.58 34.23 2.76 8.06 -0.040 28.16 34.37 39.33
digit 5
Second joint to
3 12.93 22.55 17.52 1.96 11.19 0.046 14.12 17.45 21.53
root digit 5
Finger tip to root
4 69.79 90.80 79.05 4.31 5.45 0.384 71.44 79.18 88.41
digit 3
First joint to root
5 43.76 60.51 52.06 3.54 6.80 0.091 45.13 52.51 59.36
digit 3
Second joint to 10.61
6 19.46 32.41 25.53 2.71 0.139 21.24 25.72 30.52
root digit 3
Breadth at tip
7 10.62 15.84 12.97 1.04 8.02 0.021 11.22 13.13 15.28
digit 5
Breadth at first
8 12.72 17.60 15.10 0.93 6.16 -0.024 13.53 15.27 17.14
joint digit 5
Breadth at
9 second joint 14.73 19.79 17.06 0.99 5.80 0.276 15.50 17.17 19.26
digit 5
Breadth at tip
10 12.85 18.56 15.79 1.12 7.09 -0.082 13.54 16.02 18.07
digit 3
Breadth at first
11 15.07 19.64 17.35 0.90 5.19 0.191 15.76 17.49 19.53
joint digit 3
Breadth at
12 second joint 17.90 22.45 20.21 0.94 4.65 0.187 18.37 20.27 22.23
digit 3
Depth at tip digit
13 9.46 13.86 11.37 0.86 7.56 0.293 10.02 11.53 13.45
5
Depth at first
14 11.22 16.31 13.70 0.99 7.23 0.078 12.06 13.78 15.84
joint digit 5
Depth at second
15 13.84 19.97 16.50 1.24 7.51 0.388 14.57 16.55 19.32
joint digit 5
Depth at tip digit
16 10.32 15.35 12.99 0.98 7.54 -0.103 11.39 13.17 14.99
3
Depth at first
17 12.83 17.85 15.51 1.13 7.29 -0.036 13.60 15.69 17.84
joint digit 3
Depth at second
18 16.53 22.30 19.08 1.13 5.92 0.310 17.40 19.18 21.47
joint digit 3
19 Grip span 82.32 114.66 98.07 6.30 6.42 -0.019 86.71 99.15 109.56
Max. breadth of
20 95.00 110.00 101.83 3.38 3.32 0.278 95.00 102.00 110.00
the hand
Breadth of the
21 78.00 92.00 84.85 2.82 3.32 0.082 80.00 85.00 92.00
knuckles
22 Hand length 170.00 202.00 185.77 6.32 3.40 0.216 175.00 187.00 201.00
23 Palm length 94.00 118.00 105.59 4.57 4.33 0.188 97.00 106.00 115.00
24 Depth of the 24.00 32.00 28.04 1.68 5.99 0.010 25.00 28.00 31.00
knuckles
Max. depth of
25 35.00 54.00 44.62 3.41 7.64 0.071 40.00 45.00 51.00
the hand
26 Fist length 89.00 113.00 100.05 4.99 4.99 0.009 92.00 101.00 110.00
First phalanx
27 60.00 74.00 65.85 2.92 4.43 0.442 62.00 66.00 72.00
digit 3 length
Fist
28 252.00 305.00 277.65 10.57 3.81 -0.093 259.00 280.00 305.00
circumference
Hand
29 225.00 265.00 243.82 8.52 3.49 -0.100 228.00 245.00 262.00
circumference
Max. hand
30 310.00 379.00 344.50 12.87 3.74 -0.251 319.00 346.00 373.00
circumference
Index finger
31 60.00 77.00 67.28 3.76 5.59 -0.075 61.00 68.00 74.00
circumference
Wrist
32 149.00 185.00 164.54 6.92 4.21 0.153 152.00 165.00 180.00
circumference
33 Arm length 692.00 847.00 771.16 27.36 3.55 -0.025 727.00 776.00 821.00
34 Elbow length 423.00 501.00 459.91 15.70 3.41 0.260 434.00 462.00 493.00
35 Elbow flexed 223.00 320.00 263.72 18.11 6.87 0.113 234.00 266.00 295.00
Max. internal
36 35.00 52.00 42.68 4.05 9.49 0.163 35.00 44.00 50.00
grip diameter
Middle finger
37 palm grip 12.00 22.50 16.33 2.47 15.12 0.188 12.50 17.50 21.00
diameter
TABLE 3: Hand Anthropometric data of sample (N=878, All measurements are in Millimeter)
In addition to the above analysis the male industrial worker groups were divided further into five
age groups of 18-25, 26-35, 36-45, 46-55, and above 56 years, for which mean and standard
deviations, were calculated separately as shown in table 4. Based on these values, the 5th, 50th
and 95th percentiles can be calculated separately.
18-25 46-55 (n= 56-Above (n=

S. Hand 26-35 (n=253) 36-45 (n=221)
(n=133) 218) 53)
No dimensions
Mean SD Mean SD Mean SD Mean SD Mean SD
Finger tip to
1 60.71 3.79 60.12 4.00 59.75 4.51 59.37 3.47 58.16 4.06
root digit 5
First joint to
2 35.49 3.05 34.78 2.48 34.13 3.37 34.06 3.47 33.35 2.78
root digit 5
Second joint to
3 18.14 2.23 17.81 1.85 17.64 2.31 17.36 1.96 16.82 2.00
root digit 5
Finger tip to
4 79.95 4.58 80.06 5.31 79.16 4.81 79.77 4.98 77.19 5.49
root digit 3
First joint to
5 52.92 3.51 53.48 3.97 51.98 4.06 52.05 3.67 49.62 4.84
root digit 3
Second joint to
6 26.14 2.62 26.25 3.07 25.52 2.38 25.93 2.63 23.49 3.24
root digit 3
Breadth at tip
7 12.82 1.19 12.87 1.26 13.20 1.20 13.69 1.09 13.28 1.25
digit 5
Breadth at first
8 15.03 1.01 15.03 1.02 15.34 1.03 15.64 1.13 15.36 1.11
joint digit 5
Breadth at
9 second joint 16.74 0.96 16.80 1.04 17.28 1.07 17.75 1.04 17.81 1.23
digit 5
Breadth at tip
10 15.36 1.38 15.76 1.49 16.12 1.18 16.50 1.26 15.94 1.14
digit 3
Breadth at first
11 17.13 0.98 17.19 1.14 17.58 1.08 18.02 1.09 17.71 1.13
joint digit 3
Breadth at
12 19.80 1.21 20.08 1.17 20.44 1.18 20.62 1.10 20.08 1.18
second joint
digit 3
Depth at tip
13 11.24 0.82 11.40 1.01 11.56 1.13 11.92 0.99 11.98 0.87
digit 5
Depth at first
14 13.40 0.93 13.61 1.00 13.84 1.42 14.41 1.04 14.01 0.89
joint digit 5
Depth at
15 second joint 16.11 1.14 16.47 1.58 16.84 1.63 17.15 1.21 16.91 1.22
digit 5
Depth at tip
16 12.76 0.97 12.88 1.13 13.25 1.03 13.52 1.08 13.38 1.09
digit 3
Depth at first
17 15.30 0.97 15.31 1.30 15.84 1.33 16.10 1.25 16.32 1.07
joint digit 3
Depth at
18 second joint 19.00 1.24 18.92 1.19 19.31 1.23 19.75 1.35 19.66 0.92
digit 3
19 Grip span 99.93 6.58 98.08 7.08 98.54 6.60 99.38 6.72 94.64 5.04
Max. breadth
20 101.41 4.51 101.85 4.11 102.68 3.80 103.54 4.26 102.26 4.30
of the hand
Breadth of the
21 84.56 3.24 85.26 3.19 85.61 3.48 86.21 3.84 85.08 3.95
knuckles
22 Hand length 186.41 8.32 187.25 8.28 188.30 7.96 188.10 7.88 182.82 7.52
23 Palm length 107.40 5.12 105.39 5.25 106.40 5.19 106.21 5.61 102.46 5.46
Depth of the
24 27.43 1.71 27.64 1.68 28.59 1.82 28.46 2.01 27.88 2.00
knuckles
Max. depth of
25 43.96 3.12 44.08 3.17 45.76 3.73 46.24 4.01 44.46 2.66
the hand
26 Fist length 99.79 5.87 100.39 6.03 101.13 5.18 101.52 4.92 99.25 4.67
First phalanx
27 66.48 3.15 66.56 3.62 66.13 2.98 66.69 2.99 65.23 3.07
digit 3 length
Fist
28 275.41 11.14 277.65 13.33 281.00 13.08 284.39 13.42 278.98 15.97
circumference
Hand
29 242.08 10.22 243.44 10.84 247.99 9.29 248.37 11.52 239.94 10.21
circumference
Max. hand
30 342.91 13.59 344.92 17.97 348.32 13.63 347.83 18.22 345.77 17.27
circumference
Index finger
31 65.52 3.90 66.27 3.74 68.82 3.34 69.59 3.92 68.24 3.34
circumference
Wrist
32 161.59 7.06 163.65 7.50 167.14 7.56 168.72 8.84 165.18 9.87
circumference
33 Arm length 772.81 30.49 777.05 33.50 773.12 31.06 773.60 27.53 768.04 27.73
34 Elbow length 463.21 17.58 462.43 20.93 462.61 16.20 463.50 16.87 456.50 13.91
35 Elbow flexed 262.79 19.78 263.30 18.24 268.10 16.29 267.75 20.13 267.64 19.92
Max. internal
36 43.82 3.78 43.80 4.13 42.16 4.84 43.44 4.46 42.16 5.03
grip diameter
Middle finger
37 palm grip 17.09 2.69 16.87 2.67 16.21 2.79 16.17 2.65 15.51 2.26
diameter
TABLE 4: Hand Anthropometric data of sample classified by Age (Mean values and standard deviation) all
measurements are in millimeter
Table 5 shows the correlation coefficients between different hand anthropometric dimensions.
These coefficients were calculated to see to what extent these dimensions are related to each
other and to what extent equipment design decisions could be based on such correlation. The
simple and multiple regression analyses were done between hand length, hand circumference
and other hand dimensions in order to find out the best set of predictors related to hand length
and hand circumference and are provided in Table 6(a) and 6(b).
4. DISCUSSIONS
From 32486 measured hand variables, 578 measured readings are rejected using stem-and-leaf
plots, histograms and box plots on SPSS software, based on the modifications of the
Kolmogorov-Smirnov and Shapiro-Wilk test as it is suitable for continuous distribution to examine
the test of normality distribution of data. Thus rejecting 1.77% (578) sample data which may due
to certain type of error while measuring the hand dimensions the result obtains indicates that the
hand variable have statistical distribution that can fit closely to normal distribution curve, as usual
in from the result of the normality test given in Table 2. These test indicates that the thirty five out
of thirty seven hand variables were normal with some deviation in other two variables, these two
variable maximum internal grip diameter and middle finger palm grip diameter are also
approximately normal (p < 0.05) knowing that a dimension is normal makes it possible to easily
derive percentiles in the distribution using the standard normal (Z) table. Otherwise, the
cumulative distribution may be used. The frequency distribution would look like a symmetrical
bell-shaped or normal curve, with most subjects having values in the mid range and with a
smaller number of subjects with high and low scores. As all the hand anthropometric dimensions
follow a normal distribution curve and errors made in using the normal distribution are either not
significant, statistically or are of little practical importance thus the probability density function of
the underlying distribution is estimated based on a sample from the population without any prior
knowledge of the mean, variance etc. of the population
Table 3 presents the summary data obtained for mean and standard deviation, as well as other
important statistical information namely minimum, maximum, and coefficient of variation,
skewness, and important percentile values for all the hand measurements of the male industrial
worker. Coefficient of variation (the ratio of standard deviation to mean) among the thirty seven
hand dimensions ranged from 3.32 to 15.12 % with 34 of them below 10% far lower than we can
assume or suggested by [35]. As the skewness of all the thirty-seven hand dimension is less than
plus or minus one (<+/- 1.0); thus hand dimension is atleast approximately normal and skewness
is not significantly different from normal, and hence we can use the mean, standard deviation and
different percentile values to easily determine the proportion of the population who fall within a
specific range of value for a given hand dimension. These values may also be used for
comparison with those published for other population.
The values of mean and standard deviation (SD) for five age groups of male industrial workers
surveyed, namely 18-25, 26-35, 36-45, 46-55, and > 56 years; pertaining to thirty seven hand
anthropometric dimensions were calculated and are presented in table 4. The data show an
increase in most hand dimensions in the middle age before declining with an increasing age. This
classification revealed that there are clear differences between the five groups. Moreover young
and middle aged worker are smaller than 56 and above age industrial worker in breadth at
second joint digit 5, depth at tip digit 5 and depth at first joint digit 3. However in other hand
dimensions, the 56 and above age industrial worker are generally smaller than both the young
and the middle aged. Figure 2 illustrate the average values obtained of hand length and hand
circumference for five different age groups. This shows that, hand length and hand circumference
vary significantly with age. These differences are very important and should be taken into
consideration in designing the hand tools or equipment that should be controlled by hands of
different age groups. [36] and many others researchers support these findings that
anthropometric data have indicated difference among age groups. It will be interesting to find out
whether these are significant difference between different age groups most of the hand
dimensions with significant differences with were not related to vertebral compression. The exact
reason for the significant differences remain unknown we could not identify them in this study.
The differences found in the hand anthropometric dimensions of the different age groups
emphasize the usefulness of this study and of the results presented herein.
Correlations among measured hand segments were performed among hand length and hand
circumference. Testing the significance of correlation revealed that almost all values are
significant and positively correlated between the hand length and hand circumference, suggest
that it is possible to predict hand dimensions with 95% confidence, by measuring the hand length
and hand circumference alone. Linear regression equations are provided in Table 6(a) and 6(b)
respectively. The statistically significant correlation between the hand lengths (L) related variables
are coded by Y1 to Y28 and the hand circumference (C) related variables are coded by Y29 to Y34.
Coefficient of
Code Variable Prediction Equation
Correlation
L Hand length - -
Y1 Finger tip to root digit 5 0.602** Y1 = 0.4346L – 20.736
Y2 First joint to root digit 5 0.486** Y2 = 0.3866L – 35.296
Y3 Second joint to root digit 5 0.299** Y3 = 0.2098L – 21.258
Y4 Finger tip to root digit 3 0.697** Y4 = 0.5322L – 18.892
Y5 First joint to root digit 3 0.610** Y5 = 0.4082L – 23.512
Y6 Second joint to root digit 3 0.470** Y6 = 0.2922L – 28.622
Y7 Breadth at tip digit 5 0.110* Y7 = 0.12L – 8.9
Y8 Breadth at first joint digit 5 0.139** Y8 = 0.1276L – 7.886
Y9 Breadth at second joint digit 5 0.181** Y9 = 0.12L – 5.09
Y10 Breadth at tip digit 3 0.038 Y10 = 0.1478L – 11.988
Y11 Breadth at first joint digit 3 0.168** Y11 = 0.125L – 6.03
Y12 Breadth at second joint digit 3 0.272** Y12 = 0.1518L – 8.458
Y13 Depth at tip digit 5 0.060 Y13 = 0.0952L – 5.792
Y14 Depth at first joint 5 0.163** Y14 = 0.1292L – 9.452
Y15 Depth at second joint 5 0.152** Y15 = 0.1746L – 14.786
Y16 Depth at tip digit 3 0.022 Y16 = 0.1188L – 8.688
Y17 Depth at first joint digit 3 0.141** Y17 = 0.1332L – 8.642
Y18 Depth at second joint digit 3 0.243** Y18 = 0.1386L – 6.046
Y19 Grip span 0.419** Y19 = 0.6674L – 24.464
Y20 Maximum breadth of the hand 0.466** Y20 = 0.46L + 17.4
Y21 Breadth of the knuckles 0.415** Y21 = 0.38L + 15.2
Y22 Palm length 0.290** Y22 = 0.6L – 6.0
Y23 Depth of knuckles 0.411** Y23 = 0.18L – 4.8
Y24 Maximum depth of hand 0.254** Y24 = 0.38L – 25.8
Y25 Fist length 0.306** Y25 = 0.58L – 5.8
Y26 First phalanx digit 3 length 0.455** Y26 = 0.34L + 4.6
Y27 Elbow length 0.607** Y27 = 1.9796L + 101.2653
Y28 Arm length 0.582** Y28 = 3.44L + 141.6
TABLE 6(a): Coefficient of Correlation between Hand Length and related variables for Haryana State
Industrial Workers and the corresponding prediction equation
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 26 27 28 29 30 31 32 33 34 35 36 37
1 1
**
2 .833 1
** **
3 .619 .752 1
** ** **
4 .649 .516 .312 1
** ** ** **
5 .553 .550 .366 .852 1
** ** ** ** **
6 .378 .384 .428 .640 .746 1
**
7 .118 -.043 .083 .089 -.009 -.006 1
** * **
8 .157 -.043 .116 .057 -.037 .040 .623 1
** ** ** ** **
9 .161 .042 .143 .174 .066 .071 .511 .684 1
- - ** ** **
10 .029 * .055 -.014 ** -.046 .556 .562 .404 1
.091 .171
* * * ** ** ** **
11 .086 .005 .098 .098 .025 .111 .435 .621 .619 .595 1
** * ** ** ** ** ** ** ** ** **
12.183 .105 .173 .276 .233 .166 .342 .468 .634 .343 .592 1
** ** ** ** ** **
13 .006 -.066 .002 .036 -.077 -.009 .516 .513 .534 .441 .480 .507 1
** ** ** ** ** ** ** ** ** ** **
14.148 .014 .132 .170 .048 .188 .416 .513 .543 .294 .417 .501 .545 1
** ** ** ** ** ** ** ** ** ** **
15.219 .069 .049 .211 .072 .141 .356 .397 .461 .350 .330 .341 .447 .605 1
- - ** ** ** ** ** ** ** ** **
16 -.050 ** -.057 .000 * .017 .444 .478 .449 .483 .460 .348 .631 .515 .487 1
.174 .093
* ** ** ** ** ** ** ** ** ** ** **
17 .051 -.061 .025 .101 .019 .157 .309 .500 .488 .318 .535 .490 .451 .596 .528 .553 1
** * ** ** ** ** ** ** ** ** ** ** ** ** ** **
18.160 .073 .091 .240 .186 .218 .234 .362 .443 .268 .517 .572 .442 .555 .502 .489 .662 1
** ** ** ** * ** *
19.179 .165 .027 .171 .126 .026 .061 -.085 -.032 -.036 .031 .093 .173 -.046 .104 .058 .009 .083 1
** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
20.304 .166 .124 .299 .180 .058 .339 .362 .355 .335 .315 .273 .291 .224 .286 .251 .257 .273 .262 1
** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
21.250 .135 .145 .389 .265 .262 .326 .366 .368 .345 .297 .338 .371 .303 .364 .302 .304 .272 .181 .668 1
** ** ** ** ** ** * ** ** ** ** ** ** ** ** ** ** **
22.602 .486 .299 .697 .610 .470 .110 .139 .181 .038 .168 .272 .060 .163 .152 .022 .141 .243 .419 .466 .415 1
** ** ** ** ** ** ** * * ** ** ** ** ** **
23.393 .305 .173 .329 .256 .229 .077 .077 .072 .000 .086 .149 .053 .115 .095 .017 .164 .158 .512 .377 .290 .772 1
** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
24.192 .073 .118 .232 .127 .135 .294 .375 .434 .292 .419 .460 .301 .318 .308 .336 .406 .374 .031 .398 .411 .254 .204 1
** ** * ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** * **
25.127 .012 .021 .146 .092 .051 .273 .301 .384 .370 .392 .398 .271 .269 .212 .216 .289 .346 .069 .397 .254 .205 .099 .411 1
** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
26.378 .294 .226 .390 .413 .365 .201 .180 .343 .045 .282 .391 .153 .351 .383 .157 .375 .408 .215 .341 .306 .569 .512 .362 .371 1
** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
27.578 .436 .278 .735 .632 .423 .180 .166 .300 .084 .220 .488 .230 .311 .266 .118 .270 .388 .191 .393 .455 .717 .466 .397 .284 .515 1
** ** ** * ** ** ** ** ** ** ** ** ** ** ** ** * ** ** ** ** ** ** ** **
28.144 -.056 .010 .319 .161 .113 .370 .370 .378 .234 .225 .290 .233 .320 .258 .245 .285 .238 .104 .619 .555 .390 .274 .449 .388 .251 .346 1
** * ** * ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
29.265 .082 .114 .269 .101 .079 .432 .427 .452 .463 .375 .383 .371 .379 .315 .314 .309 .307 .125 .770 .649 .388 .278 .525 .596 .372 .394 .682 1
** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
30.388 .202 .137 .327 .219 .142 .322 .222 .238 .252 .224 .246 .227 .244 .199 .167 .143 .182 .269 .584 .381 .503 .400 .386 .435 .398 .408 .510 .697 1
** * ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
31.152 .031 .095 .207 .060 .046 .369 .448 .567 .366 .416 .494 .422 .513 .423 .352 .455 .406 .083 .467 .463 .242 .172 .549 .547 .391 .374 .496 .631 .354 1
** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** * ** ** ** ** ** ** ** **
32.256 .085 .156 .288 .135 .124 .315 .364 .513 .328 .330 .456 .277 .409 .328 .224 .290 .359 -.043 .561 .506 .284 .093 .473 .486 .301 .373 .592 .709 .450 .603 1
** ** ** ** ** ** * * * * ** * ** * ** ** ** ** ** ** ** ** ** ** ** ** ** * **
33.471 .298 .160 .545 .434 .298 .105 .101 .114 .035 .107 .138 .043 .096 .159 .056 .103 .194 .303 .382 .365 .607 .440 .168 .202 .295 .514 .309 .391 .436 .100 .294 1
** ** ** ** * ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
34.255 .133 .124 .238 .117 -.028 .214 .269 .413 .197 .277 .395 .242 .264 .156 .242 .216 .257 .122 .533 .402 .286 .223 .363 .418 .251 .343 .439 .579 .331 .395 .579 .366 1
** ** ** ** ** ** - - - - ** ** ** ** ** ** ** ** ** ** ** **
35.408 .390 .191 .517 .475 .270 -.010 -.056 -.044 ** -.021 .074 -.069 * -.008 ** ** .016 .383 .318 .153 .585 .365 .013 .041 .216 .359 .185 .121 .334 -.047 .133 .446 .052 1
.131 .098 .175 .167
** ** ** ** ** - - * ** ** ** ** ** * ** * ** ** **
36.290 .227 .034 .371 .306 .160 -.004 -.051 -.019 * * .080 .024 .018 .099 -.017 -.002 .032 .399 0.181 .120 .400 .233 -.031 .052 .112 .275 0.103 .046 0.231 .039 .080 .248 .017 .681 1
.111 .103
** ** ** ** ** ** * ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** ** **
37.481 .297 .179 .549 .442 .351 .110 .071 .185 -.025 .088 .237 .048 .143 .219 .014 .153 .251 .264 .347 .320 .582 .463 .226 .206 .432 .580 .279 .382 .424 .158 .280 .799 .336 .369 .220 1
TABLE 5: Matrix of the Pearson Correlation Coefficients obtained between the different Hand Anthropometric dimensions as per order provided In Figure 1 and
Table 3
**Correlation is significant at the 0.01 level (2- tailed) *Correlation is significant at the 0.05 level (2- tailed)
Arunesh Chandra, Pankaj Chandna & Suriendra Deswal
Coefficient of
Code Variable Prediction Equation
Correlation
C Hand circumference - -
Y29 Maximum hand circumference 0.510** Y29 = 1.6552C – 65.1724
Y30 Index finger circumference 0.496** Y30 = 0.3621C – 19.569
Y31 Wrist circumference 0.509** Y31 = 0.8448C – 38.3276
Y32 Elbow flexed 0.391** Y32 = 1.7414C – 158.8793
Y33 Maximum internal grip diameter 0.121** Y33 = 0.431C – 63.5345
Y34 Middle finger grip diameter 0.046 Y34 = 0.181C – 27.2845
TABLE 6(b): Coefficient of Correlation between Hand Circumference and related variables for Haryana State
Industrial Workers and the corresponding prediction equation
** Significant at α = 0.01 * Significant at α = 0.05

Note all dimensions in mm
The tests of hypothesis that the intercepts or the slopes are zero were rejected for the level of
significance shown in Table 6(a) and 6(b) Predictions should be confined to the ranges of hand
length and hand circumference as prescribed by the regression analysis. The minimum and
maximum values for hand length were 170 mm and 202 mm respectively and the counter values for
the hand circumference were 225 mm and 244 mm respectively. Although this hand anthropometric
data will be of great value in practical application it should be noticed that these are static
anthropometric measurers. Therefore, the use of such data in design of equipment, tools, and
workstation in which functional hand anthropometric data is needed, must be done considering the
differences between the two referred types of hand anthropometric data
250
189
248
Hand Circumference (mm)
188
246
187 244
186 242
185 240
Hand length (mm)
238
184
236
183
234
182 18-25 26-35 36-45 46-55 56-Above
181 Age group (yrs)
180
18-25 26-35 36-45 46-55 56-Above
Age group (yrs)
FIGURE 2: Variation of Hand Length and Hand Circumference (Mean Values in mm) for Age Groups (Yrs.)
defined
5. CONCLUSIONS
Thirty-seven hand dimensions of eight hundred and seventy eight male industrial workers of
Haryana state belonging to thirty-eight industries of Haryana state of India have been analysed in
this work. This will be useful for the new designs/design modifications for hand tools, workstations,
hand apparel, tools and protective equipment and other practical applications. Mean and standard
deviation of the sample of different age groups shows that values of most of the hand
anthropometric dimensions are higher in the middle age groups and lower with higher and lower
age groups. With respect to the above analysis there are a few important remarks, which need to be
emphasized.
• This study investigated assumptions of normality commonly made by designers in
establishing workplace, equipment, as well as tool design recommendations and the
objective of this analysis is to check precision in anthropometric measures. It was observed
that 98.23% of collected reading of 37 hand variables of hand anthropometric dimensions fit
closely to a normal distribution curve.
• The correlation coefficients among different hand dimensions were calculated to see to
what extent these dimensions are related to each other. It was observed that 77% of
correlation coefficients are significant at the 1% level, 5% of the correlation coefficients are
significant at 5% level, and 18% of the remaining values are insignificant. Correlation
among measured hand segments was performed among hand length and hand
circumference and almost all values are significant and positively correlated.
• The sample size used (878) was satisfactory for all variables. Therefore designers for
industrial worker of Haryana state can utilize the statistics presented and prediction
equations present in this study to set specifications for the system used, such as hand tools
and other hand held devices. These prediction equations can be used to predict 34 hand
variable dimensions with 95% confidence by measuring the hand length and hand
circumference alone.
• There is a need to enlarge the sample size, not only in terms of age range, namely to
compensate for low frequency observed below 25 and above 56 years, but also to
encompass other occupational groups such as agricultural worker, household worker,
constructional workers and of female workers as their numbers are increasing day to day in
the state.
6. REFERENCES
[1] B. Mebarki and B.T. Davies. “Anthropometry of Algerian Woman”. International Journal of
Industrial Ergonomics 33(12): 1537-1547, 1990.
[2] W.G. Lewis and C.V. Narayan. “Design and sizing of ergonomic handles for hand tools”.
Applied Ergonomics 24: 351-356, 1993.
[3] W.E. Botha and R.S. Bridger. “Anthropometric variability, equipment, usability and
musculoskeletal pain in a group of nurses in the Western Cape”. International Journal of
Applied Ergonomics 29(6): 481-490, 1998.
[4] P. Chandna, S. Deswal and A.Chandra. “An Anthropometric Survey of Industrial Workers of
the Northern Region of India” International Journal of Industrial and Systems Engineering
6(1): 110-128, 2010.
[5] G.C. Marks, J.P. Habicht and W.H. Mueller. “Reliability, dependability and precision of
anthropometric measurement”. American Journal of Epidemiology 130(3): 578-587, 1989.
[6] B. Buchholz and T.J. Armstrong. “An ellipsoidal representation of human hand anthropometry”.
Human Factors 33: 429-441, 1991.
[7] K.N. An, E.Y. Chao, W.P. Cooney and R.L. Linscheid. “Normative model of human hand for
biomechanical analysis”. Journal of Biomechanics 12: 775-788, 1979.
[8] B. Buchholz, T.J. Armstrong and S.A. Goldstein. “Anthropometric data for describing the
kinematics of human hand”. Ergonomics 35: 261-273, 1992.
[9] O.O. Okunribido and K.A. Olajire. “A survey of hand anthropometry of female rural workers in
Ibadan, Western Nigeria”. Ergonomics 11: 2-6, 1999.
[10] M.H. Yun. “Designing for diversity”. In Proceedings of the human factors and ergonomics
society. 37th Annual Meeting, Seattle, Washington, 1993.
[11] C. Frannson and J. Winkel. “Hand strength: the influence of grip spans and grip type”.
Ergonomics 34: 881-892, 1991.
[12] J.R. Blackwell, K.W. Kornatz and E.M. Heath. “Effect of grip span on maximal grip force and
fatigue of flexor digitorum superficial”. Applied Ergonomics 30: 401-405, 1999.
[13] C.B. Irwin and R.G. Radwin. “A new method for estimating hand internal loads from external
force measurements”. Ergonomics 51(2): 156-167, 2008.
[14] S.N. Imrhan and K. Farahmand. “Male torque strength in simulated oil rig tasks and the effects
of grease-smeared gloves and handle length, diameter and orientation”. Applied Ergonomics
30(5): 455-462, 1999.
[15] K.A. Grant, D.J. Habes and L.L. Steward. “An analysis of hand designs for reducing manual
effort: The influence of grip diameter”. International Journal of Industrial Ergonomics 10(3):
199-206, 1992.
[16] M. Eksioglu, J.E. Fernandez and J.M. Twomey. “Predicting peak pinch strength: artificial
neural networks vs. regression”. International Journal of Industrial Ergonomics 18: 431-441,
2004.
[17] N.A. Snow and T.J. Newby. “Ergonomically designed job aids”. Performance and Instructional
Journal 28: 26-30, 1984.
[18] A. Chandra, P. Chandna, S. Deswal and R. Kumar. “Ergonomics in the Office Environment: A
review”. In Proceedings of the International Conference on Energy and Environment.
Chandigarh, Haryana, India, 2009.
[19] L. Claudon. “Ergonomics hand tools design: Interview of users. Ergonomics and safety for
global business quality and productivity”. In Proceeding of the second international
conference ERGON-AXIA 2000. Warsaw, Poland. Central Institute for Labour Protection;
Warsaw. (2000),
[20] P. Loslever and A. Ranaivosoa. “Biomechanical and epideminological investigation of carpal

tunnel syndrome at work-places with high risk factors”. Ergonomics 36(5): 537-555, 1993.
[21] S.N. Imrhan, M. Nguyen and N. Nguyen. “Hand anthropometry of Americans of Vietnamese
origin”. International Journal of Industrial Ergonomics 12: 281-287, 1993.
[22] J.D.A. Abeysekera and H. Shahnavaz. “Body size variability between people in developed
and developing countries and its impact on the use of imported goods”. International Journal
of Industrial Ergonomics 4: 139-149, 1989.
[23] O. Okunribido. “A survey of hand anthropometry female rural farm workers in Ibadan, Western
Nigeria”. Ergonomics 43: 282-292, 2000.
[24] S.K. Kar, S. Ghosh, I. Manna, S. Banerjee and P. Dhara. “An investigation of hand
anthropometry of agricultural workers”. Journal of Human Ergology 14(1): 57-62, 2003.
[25] L.T. Gite and B. G. Yadav. “Anthropometric survey for agricultural machinery design: An
Indian Case Study”. Applied Ergonomics 20: 191-196, 1989.
[26] A. Nag, P.K. Nag and H. Desai. “Hand anthropometry of Indian women”. Indian Journal
Medical Research 117: 260-269, 2003.
[27] S.N. Imrhan and M.G. Contreras. “Hand anthropometry in a sample of Mexicans in the US-
Mexico border region”. In Proceedings of the XIX annual Occupational ergonomics and safety
conference, Las Vegas, NV: 589-593, 2005.
[28] N. Mandahawi, S. Imrhan, S. Al-Shobaki and B. Sarder. “Hand anthropometry survey for the
Jordanian population”. International Journal of Industrial Ergonomics 38: 966-976, 2008.
[29] S.N. Imrhan, M.D. Sarder and N. Mandaharu. “Hand anthropometry in Bangladeshis living in
America and comparisons with other populations”. Ergonomics 52: 987-998, 2009.
[30] A. Chandra, P. Chandna and S. Deswal. “Hand Anthropometric Survey of Male Industrial
Workers of Haryana State (India)”. International Journal of Industrial and Systems
Engineering (In Press 2011).
[31] B.T. Davies. “Female hand dimensions and guarding of machines”. Ergonomics 23(1): 79-84,
1980.
[32] A.J. Courtney and M.K. Ng. “Hong Kong female hand dimensions and machine guarding”.
Ergonomics 27(2): 187-193, 1984.
[33] H. Kemper and J. Pieters. “Comparative study of anthropometric measurements of the same
subjects in two different institutes”. American Journal of Physical Anthropology 40: 340-344,
1974.
[34] E. Panchon, R. Lobato, F. Sanchez and A. Panchon. “Index for quality control in
anthropometric surveys”. International Journal of Industrial Ergonomics 34: 479-482, 2004.
[35] S. Pheasant. “Bodyspace: Anthropometry, Ergonomics and the Design of Work”, Taylor &
Francis, (1998).
[36] L.G. Martin and B.J. Soldo. “Racial and Ethnic Differences in the Health of Older Americans”,
National Academy Press, Washington, DC, 285-296, 1997.
Mohammed S. Al-Numay & N.M. Adamali Shah
Averaging Method for PWM DC-DC Converters Operating in

Discontinuous Conduction Mode With Feedback
Mohammed S. Al-Numay [email protected]

Department of Electrical Engineering
KingSaudUniversity,
Riyadh11421, Saudi Arabia.
N.M. Adamali Shah [email protected]

Department of Electrical Engineering
KingSaudUniversity,
Riyadh11421, Saudi Arabia.
Abstract
In this paper, a one-cycle-average (OCA) discrete-time model for PWM dc-dc converter based
onclosed-loop control method for discontinuous conduction mode (DCM) is presented. It leads
toexact discrete-time mathematical representation of the OCA values of the output signal even
atlow frequency. It also provides the exact discrete-time mathematical representation of
theaverage values of other internal signals with little increase in simulation time. A comparison of
thismodel to other existing models is presented through a numerical example of boost
converter.Detailed simulation results confirm the better accuracy and speed of the proposed
model.
Keywords: Switched Systems, Pulse-width Modulation (PWM), Power Converter, Discrete

TimeModeling, one-cycle-averaging, Sampled Data Model.
1. INTRODUCTION
PWM converters are widely used for operating switch controlled systems. These systems
areusually operated in two modes of operation, namely: continuous and discontinuous
conductionmodes [5]. The discontinuous conduction mode (DCM) of operation typically occurs in
dc/dcconverters at light load.The boundary between the continuous conduction mode (CCM)
andDCM depends on the ripple current in the inductor or the ripple voltage in the capacitor.
Forlowpower applications, many designers prefer to operate in the DCM in order to avoid the
reverserecovery problem of the diode. DCM operation has also been considered a possible
solution tothe right-half plane (RHP) zero problem encountered in buck-boost and boost derived
topologies. In single-phase ac/dc converters with active power factor correction (PFC), the input
inductorcurrent becomes discontinuous in the vicinity of the voltage zero crossing; some PFC
circuits areeven purposely designed to operate in DCM over the entire line cycle in order to
simplify thecontrol. Proper analytical models for DCM operation of PWM converters are therefore
essentialfor the analysis and design of converters in a variety of applications (see [11] and
referencestherein). These modes of operation are also very much useful for efficiently extracting
maximumpower from the photovoltaic panel (PV) which is another main application [12]. These
powerconverters are connected between the PV and load or bus. Due to the variety of
applications ofPWM converters operating in DCM, there is a need for an accurate model for the
analysis anddesign of such converters. Many efforts have been taken in this view for past three
decades [8,11].
Most power electronic design procedures rely on averaging techniques. Averaging

techniquesprovide the analytical foundation for most power electronic design procedures of the
system level. These techniques are widely used in the analysis and control design of PWM power
electronicsystem. In fact classical averaging theory is not applicable when there are state
discontinuities.This is significant because all feedback controller converters are state

discontinuous systems.
FIGURE 1: Closed loop Boost converter
Theperiodic solution of PWM converters is averaged to equilibrium. Although the periodic

solutionat high switching operations has small amplitude (ripple), and averaging seems justified,
theinherent dynamics for a periodic solution and at equilibrium are completely different. This
issueis generally neglected in most power electronics literature. It is known that the averaged
modelsare approximate by design [4, 9]. Moreover, it has been found that the directly obtained
averagedmodels are inaccurate for converters operated in discontinuous conduction mode
(DCM). Thisleads to the need of more efforts to obtain more accurate averaging
methods.Averaging methods are sometimes used to produce approximate continuous-time
models forPWM systems by neglecting the switching period of the switches and the sampling
period ofthe controller [1, 2]. In averaging process the ripple in the current or in the voltage is also
notconsidered. To overcome the above disadvantage, the sampled-data modeling techniques
areadapted. This provides the most accurate result, which replicates the actual behavior of
PWMsystems and is also suitable for digital control process. Sampled-data models allow us to
focuson cycle-to-cycle behavior, ignoring intra cycle ripples. This makes them effective in
generalsimulation, analysis and design. These models predict the values of signals at the
beginningof each switching period, which most of the times represent peaks or valleys of the
signals ratherthan average values. To better understand the average behavior of the system, a
discrete-timemodel for the OCA signals was presented in [1]. Averaged discrete-time models for
continuousconduction mode (CCM) and discontinuous conduction mode (DCM) without PWM
feedbackcontrol are available. They are shown to be more accurate than continuous-time models
andfaster [1, 2].
In this paper, a sampled-data model for PWM converters operating in DCM with feedback
isformulated. This gives the exact discrete-time mathematical representation of the values of
theoutput and internal signals with feedback loop at low frequency. A discrete-time model to
providethe one-cycle-average (OCA) signals of PWM converters operating in DCM with feedback
is proposed.This model provides the exact discrete-time mathematical representation of the
averagedvalues of the output signal. It also provides the average values of other internal signals
with littleincrease in simulation time. The main motivation for the new model is based on the fact
that, inmany power electronic applications, it is the average values of the voltage and current
rather thantheir instantaneous values that are of greatest interest. Numerical simulations show
the accuracyof the propose model.
2. EXISTING AVERAGE MODELS

Different averaging methods for PWM converters are used for comparison, analysis and
design.The mathematical models of the boost converter with PWM feedback control shown in
Figure 1are presented in this section and will be used in Numerical example.
2.1 Switched Model
, τ
The DCM PWM converter can be described by
, τ (1)

, τ
, τ
, τ

, τ
(2)
m n p
Where u∈R is the input vector, x∈R is the state vector, and y∈R is the output vector.
Thesystem switches between three topologies (A1, B1, C1), (A2, B2, C2), and (A3, B3, C3),
withswitching intervals determined by
τ
τ
τ
1 2
Where T is the switch period, (dk +dk ) ∈ [0, 1] are the switch duty ratios, and k is the discrete-
time index. All auxiliary inputs will be assumed to be piecewise constants, i.e. u(t) = uk for all t ∈
[kT, (k+1)T]. This assumption is not necessary and is made for convenience only; more general
cases would only require more complex notations. This is the exact switching model which will be
used as the base model for comparison of different methods.
The state space matrices A1,A2,A3,B1,B2,B3,C1,C2, and C3of boost converter shown in Figure1 are
0 0 1
defined as
1 !; #! ; $0 1%
0
0
1
0 1
# '; #! ; $0
&
1 1 1%
0

0 0
1 !; 0
( ) ; $0 1%
0
0

The control scheme given in is applied, where the modulation signal is m(t)=Vref- k1 i(t) - k2 v(t)
with Vref=0.13, k1 =0.174, and k2 = -0.0435 as in [1, 7].
2.2 DCM State-space Average Model (SSA)

The SSA mathematical model of the boost converter is given as [11]
.1 1
*+ */ ,*/
,- 0 2
(3)
43/
*+ ,/ -0 *
3 23
(4)
2.3 Conventional Discrete-Time Model (CDTM)

The conventional discrete-time mode is given by
5 , , (5)
Where the input nonlinearities A (d 1, d 2)and B (d 1, d 2) are given by
, Φ Φ Φ
, Φ Φ Г Г Г
The arguments d1T, d2T, and (1-d1-d2)T for (Φ1andΓ1), (Φ 2andΓ2) and (Φ3andΓ3) respectively are
omitted from the above equations for notation simplicity. Where
Φ8 е :; <
<
Г8 = > :; τ @A τ
?
3. PROPOSED MODEL
This section introduces the new average discrete-time model for PWM converter operating
inDCM with feedback loop. Description of the original system and derivation of the proposed
modelare discussed here. The one-cycle average (OCA) representation of the output signal [1] is
B C - D<F- EE

<
givenby
(6)
The signal, y*(t) is used to develop a new discrete-time model for PWM converters operating
inDCM. This model provides the basis for discrete-time simulation of the averaged value of
anystate in the DCM PWM system, even during transient non-periodic operating conditions.
3.1 OCA Discrete-Time Model

It is desired to compute, without approximation, the evolution of all system variables at
thesampling instants, t = kT assuming three different topologies for the system. Since the state
andoutput equations (1) - (2) are piecewise-linear with respect to time t, the desired discrete-time
* *
model can be obtained symbolically. Using the notation, xk := x(kT) and yk :=y (kT), the result is
the OCA large signal model
5 , ,

5
B
, G ,
(7)
(8)
Where the input nonlinearities A(d 1, d 2), B(d 1, d 2), C(d 1, d 2)and D(d 1, d 2) are given by
, Φ Φ Φ
, Φ Φ Г Г Г
, ΦB ΦB Φ ΦB Φ Φ
G , ГB ΦB Г ГB ΦB Φ Г Г ГB
The arguments d1T, d2T, and (1 - d1- d2)T for (Φ1, Φ1*,Γ1, andΓ1*), (Φ 2, Φ2*,Γ2, andΓ2*) and (Φ 3,
* *
Φ3 ,Γ3, andΓ3 ) respectively are omitted from the above equations for notation simplicity. Where
ΦA t C > :; <

<
ΓA t C = > :; J @A E
?
<
1
ΦAB t C = ΦA EE

?
<
1
ΓAB t C = ΓA EE

?
Note that the averaging operation adds “sensor” dynamics to the system;as a consequence,
thelarge-signal model equations (7) and (8) is not in standard state-space form. By defining
theaugmented state vector x* ∈Rn+psuch that
5
5
B
C K , L
G ,
An equivalent (but standard form) representation of the OCA large-signal model is given by:
5
B
B , B
,

B B B
, 0N O P
Where
B , M Q
, 0P O P
B ,
, K L

G ,
B , $0P O N RN O P %
Note that not only the OCA values of output signal will be available but also the values of
thesignals (without averaging) at the beginning of every switching period as well.
FIGURE 2: PWM sawtooth function
3.2 Feedback Computation

The modulation signal for feedback control is m(t)=Vref-k1 i(t) - k2 v(t) = Vref- K x(t) and the duty
* *
ratio at each switching period is dk=t /T. The time instant t at which the modulation signal crosses
the sawtooth is computed by solving the nonlinear equation
ST B , UVWX Y B

UVWX YZΦ B Γ B [ (9)
at each time instant k, where the sawtooth function is shown in Fig. 2 and
mathematicallyrepresented by tri(t, T)= (t/T)-floor(t/T). For reasonably high switching frequency,
*
the value ofx(kT+t ) can be approximated by neglecting the higher order terms in the Taylor
expansion of the nonlinear functions Φ1andΓ1. That is
B/
Φ B R B ^_R B
2!
B
/
Г B R B
^ _ R B
2!
And hence, a good approximation of (9) becomes
ST B , _ UVWX YZR B B [
Noting that tri(t*,T) equals to t*/T for t*∈ [kT, (k+1)T], we get
B
_ UVWX YZR B B [

Or
B UVWX Y
_
Y 1
Which provides a closed from solution for dk. The duty ratio dk can be computed without
approximation by solving the nonlinear equation (9) for t*.
4. NUMERICAL EXAMPLE
Since all of the aforementioned averaged models have been controlled with the same
controllerdesign, a comparative study is carried out to investigate the accuracy and speed of the
proposedmodel as compared to the existing averaged model. It should be noted that no
approximation ismade in deriving the new discrete-time model, and all simulations were
performed using Matlab.The results of all models are computed using built-in Matlab nonlinear
equation solver. The state variables are x1= iLand x2= vC.
4.1 Ideal Condition

Ideal boost converter with PWM feedback control shown in Figure 1 is simulated using
existingaveraged models and proposed model. The simulated voltage and current waveforms are
shownin Figures 3 - 4. The steady-state average values predicted by the proposed model are
moreaccurate than the ones obtained by the SSA models for the parameters R = 45 Ω, L = 100
µH,C = 4.4 µF, Vg= 5 V, and TS= 100 µs.
The steady-state average values of the output voltage are v C= 8.1125 V for SSA model andvC=
8.3174 V for the proposed model. It should be noted that the accuracy of the SSA
methoddecreases as the switching frequency decreases, while the proposed model does not
depend onthe switching frequency as discussed below.
FIGURE 3: Simulation comparison of voltage waveform
4.2 Effect of Switching Frequency

To demonstrate the effect of switching frequency, consider operating the converter in a
Higherfrequency, for example TS = 80µs i.e. fS= 12.5 KHz. Figures 5 - 6 shows the effect
ofincreasing the switching frequency on the simulation results of the output waveform. The
steadystateaverage values of the output voltage are vC= 8.055 V for SSA model and vC= 8.475V
for the proposed model. The accuracy of SSA averaged model decreases and move awayfrom
the actual average as the switching frequency increases. The proposed model captures thecycle-
to-cycle behavior of the system with no approximation regardless of changes in
switchingfrequency.
FIGURE 4: Simulation comparison of current waveform
FIGURE 5: Simulation comparison of voltage waveform with higher frequency
FIGURE 6: Simulation comparison of current waveform with higher frequency
4.3 Effect of Change in load

To study the effect of load resistance on the simulation results, a step change on the
loadresistance R from 45 Ωto 55 Ωat time instant, t = 0.6 ms has been simulated and the
resultsare shown in Figure 7. The steady-state average values of the output voltage are vC=
8.1125V and v C= 9.1125 V for SSA model and vC= 8.3174 V and vC= 9.625 V for the
proposedmodel respectively. It can be observed that the average values produce by SSA model
deviatemore from the exact waveforms as the load resistance increased. On the other hand the
proposedmodel provides the same accuracy of waveforms regardless of the change in load
resistance.
Table 1 summarizes the normalized simulation times for ideal boost converter with feedback
fordifferent simulation methods.
Normalized
Method
Simulation Time
Switched 42
SSA 1
CDTM 8.4
DCM OCA 8.81
TABLE 1: Simulation time for boost converter in DCM with feedback
FIGURE 7: Simulation comparison of voltage waveform with variable load resistance
5. CONCLUSION
This paper proposed a new model which provides a discrete-time response of the one-cycle-
average(OCA) value of the output signal for PWM dc-dc converters operating in the DCM
withfeedback. It is compared to existing models through a numerical example of boost converter.
Asa result of variations in the circuit parameters such as switching frequency and load
resistance,a significant deviation in the average values of the converter’s signals is predicted by
the existingaveraging method. On the other hand, the proposed model is fast and can accurately
simulatethe average behavior of the output voltage even though there is a large variation in the
circuitparameters without any approximation in the design.
6. REFERENCES
[1] M. S. Al-Numay. “Discrete-Time Averaging of PWM DC-DC converters with feedback”, The
2006International Conference on Computer Engineering and Systems, pp. 45–48,November
2006.
[2] M. S. Al-Numay. “Averaged Discrete-Time Model for Fast simulation of PWM Converters in
DCM”,International Journal of Modeling and Simulation, vol. 29, no. 1, pp. 46–51, 2009.
[3] V. A. Caliskan, G. C. Verghese, and A. M.Stankovic.“Multifrequency averaging of DC/DC

converters”,IEEE Transactions on Power Electronics, vol. 14, no. 1, pp. 124–133, 1999.
[4] A. Davoudi, J. Jatskevich, and T. D. Rybel.“Numerical state-space average-values modeling

OfPWM DC-DC Converters Operating in DCM and CCM”, IEEE Transactions on Power
Electronics,vol. 21, no. 4, pp. 1003-1012, July 2006.
[5] N. Femia and V. Tucci.“On the Modeling of PWM Converters for Large Signal Analysis in
Discontinuous Conduction Mode”, IEEE Transactions on Power Electronics, vol. 9, no. 5, pp.
487–496, September 1994.
[6] J. G. Kassakian, M. F. Schlecht, and G. C. Verghese: Principles of Power Electronics

Reading,MA: Addison Wesley, 1991.
[7] B. Lehman, and R. M. Bass. “Switching Frequency Dependent Averaged Models for PWM
DCDCConverters”, IEEE Transactions on Power Electronics, vol. 11, no. 1, pp. 89-
98,January1996.
[8] D. Maksimovic and S. Cuk. “A Unified Analysis of PWM Converters in Discontinuous

Modes”,
IEEE Transactions on Power Electronics, vol. 6, no. 3, pp. 476–490, July 1991.
[9] R. D. Middlebrook and S. Cuk. “A general unified approach to modeling switching-converter

Powerstages”, International Journal of Electronics, vol. 42, pp. 521–550, 1977.
[10] S. R. Sanders, J. M. Noworolski, X. Z. Liu, and G. C. Verghese. “Generalized Averaging

Methodfor Power Conversion Circuits”, IEEE Transactions on Power Electronics, vol. 6, no.
2, pp. 251–259, April 1991.
[11] J. Sun, D. M. Mitchell, M. F. Greuel, P. T. Krein and R. M. Bass. “Averaged Modeling of

PWMConverters Operating in Discontinuous Conduction Mode”, IEEE Transactions on
Power Electronics,vol. 16, pp. 482–492, April 2001.
[12] K. K. Tse, M. T. Ho, Henry S. H. Chung, and S. Y. (Ron) Hui. “A Novel Maximum Power
Point Tracker for PV Panels sing Switching Frequency Modulation”, IEEE Transactions on
PowerElectronics, vol. 17, no. 6, pp. 980–989, November 2002.
[13] G. C. Verghese, M. E. Elbuluk and J. G. Kassakian, “A general approach to sampled-data

modeling for power electronics circuits ”, IEEE Transactions on Power Electronics, vol.
1(2),
pp.76–89, 1986.
[14] R. C. Wong, H. A. Owen and T. G. Wilson,“An efficient algorithm for the time-domain
Simulationof regulated energy-storage dc-to-dc converters”, IEEE Transactions on Power
Electronics, vol.2(2), pp. 154–168, 1987.
Manoj Manuja & Deepak Garg
Semantic Web Mining of Un-structured Data: Challenges and

Opportunities
Manoj Manuja [email protected]

Principal, Education & Research Dept,
Infosys Technologies Ltd.,
Chandigarh, India
Deepak Garg [email protected]

Thapar University,
Patiala, India
Abstract
The management of unstructured data is acknowledged as one of the most critical unsolved
problems in data management and business intelligence fields in current times. The major reason
for this unresolved problem is primarily because of the actuality that the methods, systems and
related tools that have established themselves so successfully converting structured information
into business intelligence, simply are ineffective when we try to implement the same on
unstructured information. New methods and approaches are very much necessary. It is a known
realism that huge amount of information is shared by the organizations across the world over the
web. It is, however, significant to observe that this information explosion across the globe has
resulted in opening a lot of new avenues to create tools for data management and business
intelligence primarily focusing on unstructured data. In this paper, we explore the challenges
being faced by information system developers during mining of unstructured data in the context of
semantic web and web mining. Opportunities in the wake of these challenges are discussed
towards the end of the paper.
Keywords: Semantic Web, Web Mining, Unstructured Data.
1. INTRODUCTION
The last few years have seen growing recognition of information as a key business tool for the
success of the organizations across the world. The organizations which effectively identify,
accumulate, study, scrutinize and thereafter act upon the information are definite winners in this
new “information age”. Further to this, the realization of “web” has critically changed the
perspective of how the organizations extract information from the available data in today’s world
of dynamic business.
Therefore, the most important differentiator between a successful and an unsuccessful business
is how an organization manages its data. The critical aspect in today’s business scenario is how
data is converted into Information and subsequently how information is converted into knowledge.
2. BUSINESS DILEMMA IN A LARGE ORGANIZATION

It is very crucial to extract knowledge from un-structured data which is available in various
formats and generated by heterogeneous sources across a big organization.
According to projections from Gartner, professionals will spend anywhere between 30 to 40 % of

their office time in managing various documents, which is 20% more than what they used to
spend on similar activities 10-15 years ago. Similarly, Merrill Lynch has anticipated that data
which is unstructured will amount to more than 85 percent of all information available in a
company.
It is very easy to extract useful knowledge from structured data using proven algorithms and
patterns. But the problem comes when we have unstructured data to work with. It becomes very
difficult to extract knowledge from the un-structured data because of non-availability of proven
algorithms, schemas, patterns and information systems. Through this paper, we shall explore
various challenges in the field of unstructured data mining using semantic web techniques and
also available opportunities in the wake of these challenges.
3. DATA, INFORMATION AND KNOWLEDGE

In a practical real life scenario, data is available in three forms:
• Unprocessed data which is gathered in real time,
• Extracted data which gives us information and
• Processed data which provides us useful knowledge [1].
Knowledge is being used to determine and analyze the specifics of a given situation. Knowledge
is a credence that is factual, vindicated, and relies on no false theories.
3.1 Un-Structured Data

It refers to computerized information that is either not having any data model or cannot be directly
used by a computer program [1]. In other words, data with some form of structure may be
characterized as unstructured if its structure does not reflect a useful schema to get a desired
processing task.
Most of the business information exists as unstructured data – commonly appearing in e-mails,
blogs, discussion forums, wikis, official memos, news, user groups, chatting scripts on social
networking sites, project reports, business proposals, public surveys, research and white papers,
marketing material, official and business presentations and most of the web pages on WWW.
3.2 Different forms of un-structured data

We have different types of un-structured data generated by different users across the globe:
• Business Data: This type of data is primarily generated in a business organization.
Although a small part of it is structured data like employee information, salary details,
company’s balance sheet etc, but a large part of it is simply unstructured like customer
communication and feedback, client presentations, minutes of project / team meeting,
official memos and many more.
• Social networking data: This type of data is purely un-structured. Users basically use
SMS type language which is not easily understandable by even human beings. Product
reviews, and feedbacks are another important part of this database. Chat scripts are also
constituents of such data.
• General communication data: This type of data mainly constitutes emails, blogs, wikis,
news, discussion forums etc. Although templates to capture this data are structured but the
contents inside those text blocks are mainly un-structured.
• Audio-Visual data: The data in the form of audio and video files is available in huge
quantity across the world. There is no defined pattern available while we browse these files.
4. CURRENT SCENARIO OF DATA MINING OVER WEB

Keeping in view of the potential opportunity to extract business focused knowledge from the
colossal amount of data available on www, a structured approach is being followed by data
administrators and managers across the world when we talk about data mining over web. Below
are the major fields which are being explored in terms of finding knowledge from un-structured
data:
4.1 Data mining with a focus towards mining unstructured data

A lot of unstructured data is noisy text [2]. Spontaneous communication (such as e-mails,
discussion forums, SMS, blogs, and collaborative web portals) contains noisy text and processing
noise. We can define “noise” in text as the difference of any type found between the original and
received text.
In the context of unstructured content [3], there is no conceptual and data type definitions
available in textual documents, and we find it very tricky to extract information from the content
[4]. Therefore, proficient algorithms duly supported by human intercessions are necessary to
make the unstructured data smoothly readable and understandable by a computer machine [5]. A
vast proportion of this unstructured data contains informal and semi-formal, internal and external
communications of a given organization [6]. Usually humans can understand such text
straightaway. However, with enormous quantity of such data content being available nowadays,
both online and inside the enterprise, it becomes critical to mine such text using computers as it
becomes very difficult and complex for a human being to mine huge data manually.
We can think of using available data mining generalized models to represent unstructured data
also but with very less efficiency and proper outcome. There are a few algorithms available to
extract useful information from unstructured data including Opinion mining [7] from noisy text
data, but a generalized, rugged approach is still missing.
4.2 Semantic Web Mining

The semantic web is based on the visualization of Tim Berners-Lee [8], the inventor of the World-
Wide-Web (WWW). According to him, “The semantic web is not at all visualized as a separate
web but it is an expansion of the existing one, in which information is given well-defined sense
and significance, better enabling PCs and people to work in cooperation.”
Knowledge Authoring Knowledge Extraction
Content Population User Query Interface
Ontology Modeling
Query Engine
Process Modeling
Rule Modeling Engine for Rule /

Reason Extraction
Domain Specific Knowledge Base
FIGURE 1: Semantic Web Solution Architecture
Semantic web mining intends at two emergent research areas of semantic web and web mining
[9,10]. The idea is to improvise the results of web mining by taking advantage of the new
semantic structures on the Web; and also, making use of web mining, for building up the
semantic web by extracting similar meanings, useful patterns, structures, and semantic relations
from existing web resources.
Figure 1 shows proposed solution architecture for semantic web mining. The architecture is
primarily divided into three logical modules, namely knowledge extraction block, knowledge
authoring block and domain specific knowledge base. The availability of prevailing search
engines has to a great extent improved our ability to carry-out a meaningful data search on the
web. But, such search option is still primarily restricted to structured data. In semantic technology,
the focus is generally to formulate flexible data model (called Triples) from the user friendly
domain query. Semantic search engines are yet to prove themselves in the huge periphery of
web search.
4.3 Ontologies Extraction

Extracting ontology from the web is a challenging task [10]. Ontology extraction and modeling use
a lot of existing resources, like text, thesauri, dictionaries, databases and similar resources.
Techniques from several related research areas e. g., machine learning, information retrieval [11],
etc, are combined, and are applied together to discover the `semantics' in the data and to make
them plain and clear.
A few systems have already been developed by research community across the world to extract
ontology [12]. Several standards have been developed to implement the layered structure of the
Semantic Web, such as the Resource Description Framework (RDF) [13] and Web Ontology
Language (OWL) [14]. Resource Description Framework (RDF) is being used by people to
represent metadata of web pages which can be processed by a machine. It describes a data
model to represent all relations between different resources. Still, this “similar meaning and
relation extraction” work is yet to mature on the global information retrieval platform simply
because of the non-availability of proven processes, standards and systems.
5. CHALLENGES IN SEMANTIC WEB MINING OF UNSTRUCTURED DATA

There are many challenges when it comes to mining of the unstructured data in the context of
semantic web:
i. Structured-data mining focused search engines: The emergence of some great
search engines has significantly improved our skillful capability to search for data on the
web; however, such search tool is vastly restricted to structured data only. Not many
search engines are available in public domain which specifically addresses the
requirement of mining / searching unstructured data flavored with semantic search. This
is the biggest challenge being faced by industry and academia alike.
ii. No standardized web form structure: We can search for and extract information
available as HTML, but till date, we are not proficient to gain easy access to the hidden
web. It is very difficult to get to the accurate web form, and even harder to find a suitable
truthful web application and related service. When we find the accurate web form or web
service, then there is a supplementary step to understand its schema and reformulate the
user's query to fit that schema. While human beings do this on a regular basis, one form
at a time, it is very complex and cumbersome to automate the process of query
reformulation, and therefore we cannot leverage the wealth of information residing behind
web forms and services for the masses.
iii. Non-availability of standard Semantics: We cannot apply the techniques for exploiting
corpora of documents directly for searching unstructured data. The main reason is that
searching unstructured data requires an understanding of its underlying semantics. This
structure is normally specified by the schema. However, in specifying these semantics,
the actual words used and the information clustering purely depends more on the
developer's whim, and little variations may result in a very different semantics altogether.
Thus, it is a big challenge to have standard semantics available for general usage.
iv. Lack of global Standards: Very less international standards are available on Semantic
Web Mining. Some big organizations and universities are working on it with little success.
Non-availability of broad, rugged and internationally recognized set of standards
addressing amalgamated mix of semantic web, web mining and unstructured data mining
is the crucial challenge being faced by researchers across the world.
v. Lack of proven frameworks: There are some challenges involved in large scale
integration on the web [15] namely the realization of the mining framework, the
robustness of mining techniques, and the exploration of holistic insight.
vi. Non-standard implementation: There is a huge implementation challenge to develop a

database management system for administering the entire process of information
extraction in an efficient and effective manner [16]. Some industries and academic
establishments have come up with their own KDIS (knowledge driven information
systems), but all these systems are proprietary to individual organizations or universities.
Because of this, the research in this area is not getting opened up for wider coverage and
business implementation.
vii. Non-standard Information Systems: If we are to design such a system which can help
the users to mine unstructured data, how should it look for? What will be the system
capabilities? The key challenges include data model and representational issues; need
for newer index structures; standardization for Information Extraction (IE); data cleaning
and its fusion; accurate relationships in the context of IE and probabilistic databases; and
lastly the role of knowledge which user possess and the iterative nature of user
interaction. These are a few challenges researchers face during information system
development.
viii. No support for audio / video data mining: There is any standard support available
which can help the end-users to extract valuable and handy information from audio and
video files available in huge quantity across the world.
ix. Less-explored Ontology framework: Developing the ontology for the large scale
databases is a great challenge in itself. There are so many industry verticals and
domains available across the organizations. Developing Ontologies for these verticals
and domains is a major challenge.
x. Lack of availability of best practices: Data vocabulary complemented by content
relevance is the operational challenges during the development of semantic web mining
tools. There are no best practices available for this development as the technology area
is not yet matured and lot of new developments are happening across the world in this
field.
6. OPPORTUNITIES FOR SEMANTIC WEB DATA MINERS IN THE FIELD OF

UN-STRUCTURED DATA
The unprecedented success of www has unfolded the true potential of two fast-emerging
research areas of semantic web and web mining. Both of these areas complement each other
and open up new opportunities for the researchers across the world. As discussed in preceding
sections, the majority of the available data on web is totally un-structured which can be
understood by human-beings only. But the amount of data suggests that the same can be
processed by machines efficiently. Hence, there is a good opportunity for semantic and web
miners to explore this situation to provide next level of mining paradigm to the world.
The intact opportunity in the field of semantic web mining can be elaborated and split into two
unique parts as “semantic” – “web mining” or “semantic web” – “mining”. In the past few years,
there have been many attempts at “breaking the syntax barrier” on the web [17].
Analysis of challenges mentioned in previous section gives us an ample opportunity to explore
semantic web mining of un-structured data and extract huge amount of knowledge available un-
tapped at www. A few opportunities are suggested below:
• To develop web mining techniques that will enable the power of www to be realized. These
constitute development of web metrics and measurements, process mining, temporal
evolution of the Web, web services optimization, fraud and threat analysis, and web mining
and privacy.
• To design and develop search engines specifically focused towards mining un-structured
databases. This is the need of the hour as the success and failure of semantic web mining of
unstructured data will primarily depend on the availability of suitable and relevant search
engines.
• To design and develop information systems for exploring unstructured data available in bulk
on web primarily extracting content, structure and usage mining. An enterprise system
integrating these three spheres of web mining is very critical to the success of this field of
research [15].
• To design and develop knowledge extraction and un-structured text processing algorithms
which are either available as proprietary algorithms with some big organizations or not
available at all for general usage and exploration.
• To design and develop models for concept and ontology extraction from unstructured data.
This extraction is very important after the enormous explosion of social networking.
• To design and develop an ontology modeling algorithm which also addresses rule and
process modeling in a particular industry vertical or domain area.
• There are not many KDIS available across the organizations which can help them feel the
pulse of their customers, employees and vendors. Mining the opinion from a huge
unstructured data available on www is one of the hottest research areas in current time.
Extracting sentiment and opinion of customers’ feedback is an exciting problem to work on.
• A few enterprises and research groups are working to make standards for semantic web, web
mining and semantic web mining. This is one of the most critical fields in today’s world which
will provide directions to the research community on semantic web mining.
• To develop a user-friendly database management system to manage the entire process of
information extraction [16]. To develop a data model which addresses representational issues
of un-structured data with newer index structures is an important unexplored field. An end-to-
end solution which may provide knowledge extraction and retrieval will be a big opportunity
for the developers to develop.
• To standardize process for Information Extraction (IE); design efficient algorithms for data
cleaning and fusion; design mechanism to find out relationships between uncertainty
management in the context of IE and probabilistic databases.
7. CRITICAL ANALYSIS
Research in the field of semantic search engines is focused on various approaches and
classification theories. Miller et al. [18] talked about Navigational Searches which points to the
classification of documents based upon the intention of the user. Mangold [19] focused on
architecture, coupling, user context, query modification, transparency, structure of ontology and
relevant technology as parameters to realize semantic search. In another critical research on
semantic search engine [20], it is pointed out that augmenting traditional keyword search with
semantic techniques is considered as the important parameters to implement the semantic
search engine.
Hildebrand et al. [21] suggested a search system based upon query construction in section with
custom search algorithms. Dietze and Schroeder [22] suggest a new classification approach
based on 9 criteria which include structured/unstructured file, text mining type, type of documents,
number of documents, Ontologies, clustering, result type, highlighting, scientifically evaluated.
Dong et al. [23] present a extended classification with semantic search algorithm based on the
Graph, methodology on distributed hash tables and logics-based Information retrieval.
8. COMPARISON OF SEMANTIC SEARCH ENGINES

In the current search scenario, there’s no denial about the super power and unquestionable
popularity of the Google search engine, where results are based on page rankings and
proprietary algorithms. But there are some very innovative ways available to search the web,
using semantic search engines. A semantic search engine will definitely ensure more closely
suggested relevant results based on the ability to understand the definition and user-specific
meaning of the word or term that is being searched for. Semantic search engines are able to
better understand the context in which the words are being used, resulting in smart, relevant
results with more user satisfaction.
A comparison of semantic search engines is shown in Table – 1.
Semantic URL Main Approach Features

Search
Engine
Kngine http://kngine.com/ Based upon "Concepts" It contains more than 8 million
concepts
Yebol http://www.yebol.com/ Based upon patented Yebol automatically clusters
algorithms paired with and categorizes search terms,
human knowledge Web sites, pages and contents,
instead of the common “listing”
of Web search queries.
Hakia http://hakia.com/ Based upon "Credible" It divides the results into Web,
News, Blogs, Twitter, Image
and Video, and can be re-listed
according to relevance.
Duckduckgo http://duckduckgo.com/ Based upon classic If we search for a term that has
search, information more than one meaning, it will
search give us the chance to choose
what you were originally looking
for, with its disambiguation
results.
EVRI http://www.evri.com/ Based upon Information Search results can be filtered
search into Articles, Quotes, Images
and Tweets.
Truevert http://www.truevert.com/ Based upon "Green All results are filtered and
Search Engine" organized from one specific
perspective – with the topic of
environmental awareness in
mind.
TABLE 1: Comparison of Semantic Search Engines
9. SOME SEMANTIC WEB BASED WEB SITES

Although, there are many web pages available on www which are using semantic web as the
base technology, we are sharing a few popular web sites as shown in Table 2.
Web Pages supported by URLs

semantic web
Brickipedia http://lego.wikia.com/wiki/LEGO_Wiki
Familypedia http://familypedia.wikia.com/wiki/Family_History_and_Genealogy_Wiki
Semantic MediaWiki http://smwtest.wikia.com/wiki/Semantic_MediaWiki_Test_Wiki
Books Wiki http://bookswiki.wikia.com/wiki/Books_Wiki
SuperWikia http://super.wikia.com/wiki/Main_Page
Governance Wiki http://government.wikia.com/wiki/Giki
Yellowikis http://yellowikis.wikia.com/wiki/Main_Page
MyWikiBiz http://mywikibiz.com/Main_Page
Common Sense Wiki http://commonsense.wikia.com/wiki/Common_Sense_Wiki
Animepedia http://anime.wikia.com/wiki/Animepedia
Creative Commons Wiki http://wiki.creativecommons.org/Main_Page
semanticweb.org http://semanticweb.org/wiki/Main_Page
TABLE 2: Semantic web based web sites
10. CONCLUSION
Semantic web mining is relatively new sub-field of data mining. It has a vast scope for
investigation keeping in view of the availability of tons of unstructured data on WWW. Lack of
available global standards on this subject opens up a enormous prospect for the research
community to focus on this area in a big way. Non-availability of a rugged database management
system to manage semantic web mining opens up new avenues for the researchers to develop
KIMS (Knowledge extraction management system) for unstructured data available on the web. A
user-oriented semantic search engine is the need of the day. These fields if explored in a right
manner will provide unlimited opportunities to extract knowledge from the goldmine of
unstructured data available across the globe.
11. REFERENCES
[1] J. Han, M. Kamber. (2001) “Data Mining Concepts and Techniques”. Academic Press,
Morgan Kaufmarm Publishers. ISBN 1-55860-489-8.
[2] W. Fan, L. Wallace, S. Rich, Z. Zhang. (2006, September). “Tapping the power of text
mining”. Communications of the ACM. Volume 49, Issue 9. pp. 76 – 82
[3] D. Bitton, F. Faerber, L. Haas, J. Shanmugasundaram. (2006). “One platform for mining
structured and unstructured data: dream or reality?”. Proceedings of the 32nd international
conference on Very large data bases. pp. 1261 – 1262
[4] R. J. Mooney, R. Bunescu. (2005). “Mining knowledge from text using information extraction”;
ACM SIGKDD Explorations Newsletter. pp. 3 – 10
[5] R. Ghani and Carlos. (2006, December). “Data mining for business applications”. KDD-2006
workshop. Volume 8, Issue 2. pp. 79 – 81
[6] M. Rajman, R. Besancon. (1997). “Text Mining - Knowledge extraction from unstructured
textual data”. In Proceedings of the 7th IFIP Working Conference on Database Semantics
(DS-7). pp. 7-10
[7] L. Dey , S. K. M. Haque. (2009, July). “Studying the effects of noisy text on text mining
applications”. Proceedings of The Third Workshop on Analytics for Noisy Unstructured Text
Data. Barcelona, Spain
[8] Tim Berners-Lee. “Semantic Web Roadmap”. http://www.W3.org/
[9] B. Berendt, A. Hotho, and G. Stumme. (2002). “Semantic Web Mining and the
Representation, Analysis, and Evolution of Web Space”. Proceedings of the First
International Semantic Web Conference on The Semantic Web. pp. 264 – 278
[10] B. Berendt, A. Hotho, G. Stumme. (2002). “Towards Semantic Web Mining”; ISWC '02:
Proceedings of the First International Semantic Web Conference on The Semantic Web;
Publisher: Springer-Verlag
[11] A Maedche. (2002). “Ontology Learning for the Semantic Web”; Kluwer. ISBN: 0792376560
[12] M. Niepert, C. Buckner, J. Murdock, C. Allen. (2008). “InPhO: a system for collaboratively
populating and extending a dynamic ontology”. Proceedings of the 8th ACM/IEEE-CS joint
conference on Digital libraries, Pittsburgh PA, PA, USA. pp. 429-429
[13] Resource Description Framework (RDF) Schema Specification. (2000) In W3C

Recommendation.
[14] Web Ontology Language (OWL), http://www.w3.org/2004/OWL/.
[15] K. Chang, B. He, Z. Zhang (2004, December). “Mining semantics for large scale integration
on the web: evidences, insights, and challenges”. ACM SIGKDD Explorations Newsletter,
Volume 6 , Issue 2. pp. 67-74.
[16] A. Doan, R. Ramakrishnan, S. Vaithyanathan. (2006). “Managing information

extraction: state of the art and research directions”. Proceedings of the ACM SIGMOD
international conference on Management of data. pp. 799 – 800
[17] G. Stummea, A. Hotho, B. Berendt. (2006). “Semantic Web Mining State of the art and
future directions”. Journal of Web Semantic. Web Semantics: Science, Services and
Agents on the World Wide Web 4. pp. 124–143.
[18] Miller, Guha, R., McCool, R. (2003). “E. Semantic Search”. Proceedings of the WWW’03,
Budapest.
[19] C Mangold (2007). “A survey and classification of semantic search approaches”.

International Journal of Metadata, Semantics and Ontologies, pp. 23–34.
[20] D. Buscaldi, P. Rosso, E. S. Arnal (2005). “A wordnet-based query expansion method for
geo-graphical information retrieval”. Working Notes for the CLEF Workshop.
[21] M. Hildebrand, J. Ossenbruggen, and L. Van Hardman (2007). “An analysis of search-
based user interaction on the semantic web”. Report, CWI, Amsterdam, Holland.
[22] F. Figueira, J. Porto de Albuquerque, A. Resende, Geus, P. Lício de Geus, G. Olso (2009).
“A visualization interface for interactive search refinement”. 3rd Annual Workshop on
Human-Computer Interaction and IR, Washington DC. pp. 46-49.
[23] H. Dong, FK Hussain, and E. Chang (2008). “A survey in semantic search technologies”.
2nd IEEE International Conference on Digital Ecosystems and Technologies.
INSTRUCTIONS TO CONTRIBUTORS
The International Journal of Engineering (IJE) is devoted in assimilating publications that

document development and research results within the broad spectrum of subfields in the
engineering sciences. The journal intends to disseminate knowledge in the various disciplines of
the engineering field from theoretical, practical and analytical research to physical implications
and theoretical or quantitative discussion intended for both academic and industrial progress.
Our intended audiences comprises of scientists, researchers, mathematicians, practicing

engineers, among others working in Engineering and welcome them to exchange and share their
expertise in their particular disciplines. We also encourage articles, interdisciplinary in nature. The
realm of International Journal of Engineering (IJE) extends, but not limited, to the following:
To build its International reputation, we are disseminating the publication information through
Google Books, Google Scholar, Directory of Open Access Journals (DOAJ), Open J Gate,
ScientificCommons, Docstoc and many more. Our International Editors are working on
establishing ISI listing and a good impact factor for IJE.
The initial efforts helped to shape the editorial policy and to sharpen the focus of the journal.
Starting with volume 5, 2011, IJE appears in more focused issues. Besides normal publications,
IJE intend to organized special issues on more focused topics. Each special issue will have a
designated editor (editors) – either member of the editorial board or another recognized specialist
in the respective field.
We are open to contributions, proposals for any topic as well as for editors and reviewers. We
understand that it is through the effort of volunteers that CSC Journals continues to grow and
flourish.
IJE LIST OF TOPICS

The realm of International Journal of Engineering (IJE) extends, but not limited, to the following:
 Aerospace Engineering  Agricultural Engineering

 Biomedical Engineering  Chemical Engineering
 Civil & Structural Engineering  Computer Engineering
 Control Systems Engineering  Education Engineering
 Electrical Engineering  Electronic Engineering
 Engineering Mathematics  Engineering Science
 Environmental Engineering  Fluid Engineering
 Geotechnical Engineering  Industrial Engineering
 Manufacturing Engineering  Materials & Technology Engineering
 Mechanical Engineering  Mineral & Mining Engineering
 Nuclear Engineering  Optical Engineering
 Petroleum Engineering  Robotics & Automation Engineering
 Telecommunications Engineering
CALL FOR PAPERS
Volume: 6 - Issue: 1 - February 2012
i. Paper Submission: November 30, 2011 ii. Author Notification: January 01, 2012
iii. Issue Publication: January / February 2012

CONTACT INFORMATION
Computer Science Journals Sdn BhD

B-5-8 Plaza Mont Kiara, Mont Kiara
50480, Kuala Lumpur, MALAYSIA
Phone: 006 03 6207 1607

006 03 2782 6991
Fax: 006 03 6207 1697
Email: [email protected]
View publication stats

IJEV5I3

Uploaded by

Copyright:

Available Formats

IJEV5I3

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

IJEV5I3

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Analysis of Hand Anthropometric Dimensions of Male Industrial Workers of

Article · August 2011

The user has requested enhancement of the downloaded file.

VOLUME 5, ISSUE 3, 2011

ISSN (Online): 1985-2312

IJE Journal is a part of CSC Publishers

Book: Volume 5, Issue 3, August 2011

IJE Journal is a part of CSC Publishers

CSC Publishers, 2011

Editorial Board Members

Dr. Kouroush Jenab

ASSOCIATE EDITORS (AEiCs)

Professor. Ernest Baafi

Dr. Tarek M. Sobh

Professor. Ziad Saghir

Professor. Ridha Gharbi

Professor. Mojtaba Azhari

Dr. Cheng-Xian (Charlie) Lin

EDITORIAL BOARD MEMBERS (EBMs)

Dr. Dhanapal Durai Dominic P

Professor. Jing Zhang

Dr. Tao Chen

Professor. Sasikumaran Sreedharan

Assistant Professor. Javad Nematian

Dr. Bonny Banerjee

AssociateProfessor. Khalifa Saif Al-Jabri

Dr. Alireza Bahadori

Professor Dr. Pukhraj Vaya

Volume 5, Issue 3, August 2011

International Journal of Engineering (IJE), Volume (5), Issue (3) : 2011

Analysis of Hand Anthropometric Dimensions of Male Industrial

Arunesh Chandra [email protected]

Pankaj Chandna [email protected]

Surinder Deswal [email protected]

Keywords: Hand Anthropometric Measurements, Industrial Worker, Hand Tools, Prediction

Male Industrial Worker

TABLE 1: Age distribution of Subjects

Kolmogorov Smirnov Shapiro-Wilk

40 Middle finger palm grip diameter 0.196 0.000 0.913 0.000

18-25 46-55 (n= 56-Above (n=

** Significant at α = 0.01 * Significant at α = 0.05

[20] P. Loslever and A. Ranaivosoa. “Biomechanical and epideminological investigation of carpal

Averaging Method for PWM DC-DC Converters Operating in

Mohammed S. Al-Numay [email protected]

N.M. Adamali Shah [email protected]

Keywords: Switched Systems, Pulse-width Modulation (PWM), Power Converter, Discrete

Most power electronic design procedures rely on averaging techniques. Averaging

discontinuities.This is significant because all feedback controller converters are state

FIGURE 1: Closed loop Boost converter

Theperiodic solution of PWM converters is averaged to equilibrium. Although the periodic

2. EXISTING AVERAGE MODELS

2.1 Switched Model

     ,  τ (1)

2.2 DCM State-space Average Model (SSA)

2.3 Conventional Discrete-Time Model (CDTM)

5   ,    ,    (5)

Where the input nonlinearities A (d 1, d 2)and B (d 1, d 2) are given by

 B  C - D<F- EE

3.1 OCA Discrete-Time Model

, τ (1)

5 , , (5)

B C - D<F- EE

5 , ,

ΦA t C > :; <

ST B , UVWX Y B

ST B , _ UVWX YZR B B [