ANALYSIS, DESIGN, AND OPERATION:ENGINEERING ANTHROPOMETRY

ENGINEERING ANTHROPOMETRY

Anthropometry is an empirical science branching from physical anthropology that deals with physical dimensions of human body and its segments, such as body size and form, including location and distribution of center of mass; segment lengths and weights; range of joint movements; and strength characteristics. Anthropometric data are fundamental to work analysis and design. Engineering an- thropometry focuses on physical measurements and applies appropriate methods to human subjects in order to develop engineering design requirements (Roebuck et al. 1975). Anthropometry is closely related to biomechanics because occupational biomechanics provides the criteria for the application of anthropometric data to the problems of workplace design (Pheasant 1989).

Anthropometry can be divided into two types: physical anthropometry, which deals with basic dimensions of the human body in standing and sitting positions (see, e.g., Tables 1 and 2), and functional anthropometry, which is task oriented. Both physical and functional anthropometry can be considered in either a static or dynamic sense. Static analysis implies that only the body segment lengths in fixed position will be considered in workplace design. Dynamic analysis requires that acceptability of design be evaluated with respect to the need to move the body from one position to another, as well as the reach and clearance considerations.

An example of the important dynamic data for workplace design is range of joint mobility (Table 3) which corresponds to postures illustrated in Figure 1. Very useful anthropometric data, both static and dynamic, are provided by the Humanscale (Henry Dreyfuss Associates 1981). When anthropometric requirements for the workplace are not met, biomechanical stresses, which may manifest themselves in postural discomfort, low back pain, and overexertion injury, are likely to occur (Grieve and Pheasant 1982). Inadequate anthropometric design can lead to machine safety hazards, loss of motion economy, and poor visibility. In other words, the consequences of anthropometric misfits may of be a biomechanical and perceptual nature, directly impacting worker safety, health, and plant productivity.

Description of Human Body Position

The anatomical body position depicts a person standing upright, with feet together, arms by the sides, and with palms forward. As a reference posture, this position is symmetrical with respect to so-called mid-sagittal plane. All planes parallel to it are also called sagittal. The vertical plane perpendicular to the sagittal is called the coronal plane. The horizontal (or transverse) plane is perpendicular to both the sagittal and coronal planes. Definition of planes of reference are especially important when the body is in other than the anatomical position.

According to Grieve and Pheasant (1982), terms of relative body position can be defined as follows. The medial and lateral positions refer to nearer to or farther from the mid-sagittal plane. The superior or inferior positions refer to nearer to or further from the top of the body. The anterior (ventral) and posterior (dorsal) positions refer to in front of or behind another structure. The super- ficial and deep positions refer to nearer to and farther from the body surface, respectively. Nearer to or farther from the trunk positions are called proximal and distal. Terms of body movements are defined in Table 4.

The Statistical Description of Anthropometric Data

The concept of normal distribution can used to describe random errors in the measurement of physical phenomena (Pheasant 1989). If the variable is normally distributed, the population may be completely described in terms of its mean (x) and its standard deviation (s), and specific percentile (Xp) values can be calculated, where: Xp x + sz, where z (the standard normal deviate) is a factor for the percentile concerned. Values of z for some commonly used percentiles (Xp) are given in Table 5. Figure 2 depicts data from Humanscale calculated for different percentiles of U.S. females. A word of caution: anthropometric data are not necessarily normally distributed in any given population (Kroemer 1989).

The Method of Design Limits

The recommendations for workplace design with respect to anthropometric criteria can be established by the principle of design for the extreme, also known as the method of limits (Pheasant 1989). The basic idea behind this concept is to establish specific boundary conditions (percentile value of the

relevant human characteristic) that, if satisfied, will also accommodate the rest of the expected user population. The NIOSH’s (1991) recommended weight limit concept is an example of application of the method of limits or design for the extreme principles to the design of manual lifting tasks. Such design is based on the expected human characteristics, where the limiting users are the weakest of the worker population.

Anthropometric Design Criteria

The basic anthropometric criteria for workplace design are clearance, reach, and posture (Pheasant 1986). Typically, clearance problems refer to design of space needed for the knees, availability of space for wrist support, or safe passageways around and between equipment. If the clearance problems are disregarded, they may lead to poor working postures and hazardous work layouts. Consideration of clearance requires designing for the largest user, typically by adapting the 95th percentile values of the relevant characteristics for male workers. Typical reach problems in industry include consid-

eration of the location of controls and accessibility of control panels in the workplace. The procedure for solving the reach problems is similar to the one used for solving the clearance problems. This time, however, the limiting user will be a smaller member of the population and the design will be usually based upon the 5th percentile value of the relevant characteristic for female workers.

Both the clearance and the reach criteria are one-tailed constraints, that is, they impose the limits in one direction only (Pheasant 1989). The clearance criterion points out when an object is too small. It does not, however, indicate when an object is too large. In some design problems, such as safe- guarding of industrial machinery, the conventional criteria of clearance and reach are often reversed.

Alternative Design Procedures

An alternative to single-percentile anthropometric design models has been presented by Robinette and McConville (1981). They point out that single-percentile models are inappropriate for both the- oretical and practical reasons. As discussed by Kroemer (1989), there are two other methods that can be used to develop the analogues of the human body for the design purposes. One method is to create models that represent the extreme ends of the body size range called the subgroup method. The other method is the regression-based procedure. which generates design values that are additive. The es- timated U.S. civilian body dimensions published by Kroemer (1981) are given in Table 1. A useful and correct general procedure for anthropometric design was recently proposed by Kroemer et al. (1986). This procedure consists of the following steps:

• Step 1: Select those anthropometric measures that directly relate to defined design dimensions. Example: hand length related to handle size.

• Step 2: For each of these pairings, determine independently whether the design must fit either only one given percentile of the body dimension, or if a range along that body dimension must be fitted. The height of a seat should be adjustable to fit persons with short and with long lower legs.

• Step 3: Combine all selected dimensions in a careful drawing, mock-up, or computer model to ascertain that all selected design values are compatible with each other. For example: the re- quired leg room clearance height needed for sitting persons with long lower legs may be very close to the height of the working surface, determined from elbow height.

• Step 4: Determine whether one design will fit all users. If not, several sizes or adjustment must be provided to fit all users.

Computer-Aided Models of Man

In order to facilitate the application of anthropometric data and biomechanical analysis in workplace design, several computer-based models of man have been developed. These computer-aided tools

should make the analysis and application of biomechanical principles at work less complicated and more useful. For a review of the state of the art in ergonomic models of anthropometry, human biomechanics and operator–equipment interfaces, see Kroemer et al. (1988). Other developments in computer-aided ergonomics, specifically computer models of man and computer-assisted workplace design, are discussed by Karwowski et al. (1990). According to Schaub and Rohmert (1990), man model development originated with SAMMIE (System for Aiding Man–Machine Interaction Evalu- ation) in England (see Figure 3) (Porter et al. 1995). Examples of computer models developed in the United States include BOEMAN (Ryan 1971) for aircraft design, COMBIMAN and CREW CHIEF (McDaniel 1990) (see Figure 4), Deneb / ERGO (Nayar 1995) and JACK (Badler et al. 1995)

Other computer-aided man models developed in Europe include ERGOMAN (France), OSCAR (Hungary), ADAPTS (Netherlands), APOLINEX (Poland), and WERNER, FRANKY, and ANY- BODY (Germany). A comprehensive 3D man model for workplace design, HEINER, was developed by Schaub and Rohmert (1990). Advances in applied artificial intelligence made it possible to develop knowledge-based expert systems for ergonomic design and analysis (Karwowski et al. 1987; Jung and Freivalds 1990). Examples of such models include SAFEWORK (Fortin et al. 1990), ERGON- EXPERT (Rombach and Laurig1990), and ERGOSPEC (Brennan et al. 1990). Other models, such as CAD-video somotograph (Bullinger and Lorenz 1990) and AutoCAD-based anthropometric design systems (Grobelny 1990), or ergonomic databases (Landau et al. 1990), were also developed.

The computer-aided systems discussed above serve the purpose of biomechanical analysis in workplace design. For example, COMBIMAN, developed in the Human Engineering Division of Armstrong Laboratory since 1975, is both illustrative and analytic software. It allows the analysis of physical accessibility (reach and fit capabilities), strength for operating controls, and visibility ac- cessibility. CREW CHIEF, a derivative of COMBIMAN, also allows the user with similar analyses. Another important development is Deneb’s ERGO, a system capable of rapid prototyping of human motion, analyzing human joint range of motion, reach, and visual accessibility. In a recent study by Schaub et al. (1997), the authors revised the models and methods of ERGOMAN and reported added capabilities to predict maximum forces / moments of relevant posture, evaluate stress of human body joints, and carry out a general risk assessment. Probably the most advanced and comprehensive computer-aided digital human model and design / evaluation system today is JACK, from Transform Technologies (2000).