Introduction to Human Factors/Ergonomics (HFE) “Engineering Anthropometry”

Introduction to Human Factors/Ergonomics (HFE)

“Engineering Anthropometry”

Hardianto Iridiastadi, Ph.D.

Introduction• Variability in physical dimensions

– Studied earlier in Anthoropology (study of mankind)

– Interest in physical aspects (beginning of anthropometry)

– Later, data are used for biomechanics investigations

• The need to design workplaces to accomodate differences in body dimensions

Human variation

Factors Affecting Anthropometrical Variation

Age Gender

Race & Ethnic Socio-economics

Occupation Life styleCircadian

Secular trend Measurement

Ergonomic Implications

• International markets– Different target countries

• Transfer of technology

• Job selection– Healthy worker effect– Fit the man to the job

Engineering Anthropometry

• “a branch of science originating from anthropology that attempts to describe the physical dimensions of the (human) body”

“anthropos” = man

“metron’ = measure

Types of Anthropometric Data

• Physical (Static) anthropometry – which addresses basic physical dimensions of the body.

• Functional anthropometry – concerned with physical dimensions of the body relevant to particular activities or tasks.

• Newtonian data – body segment mass data and data about forces that can be exerted in different tasks/postures

• Tools design

• Consumer product design

• Workplace design

• Interior design

Applications

Applied Anthropometry

Measurement Techniques

• Positions– Standing naturally upright– Standing stretched to maximum height– Lean against a wall– Sitting upright– Lying (supine posture)– “Anatomical position” (see Kroemer et al)

Measurement Techniques

• Some key measurement terms– Height– Breadth– Depth– Distance– Curvature– Circumference– Reach

Measuring Devices

• Photograph– Use of grids– Image processing techniques– Can record all three dimensional aspects– Infinite number of measurements– Drawbacks

• Parallax

• Body landmarks cannot be palpated

Newer Measuring Devices

• Whole body scanner– Ergonomic center UI– $50,000 - $400,000– Hundreds of variables– Standing and

seated posture– Combined with

modeling software

(Jack, Mannequin, etc.)

Newer Measuring Devices

Sample Anthropometric Data

Statistics

• Coefficient of variation– Data diversity = sd/mean– CV ~ 5% (10% for strength data)– Large CV should be suspected

• Standard error of the mean (se)– se = sd/√n– Useful for describing confidence interval– E.g., 95% CI = mean ± 1.96 se

• Means () and standard deviations () are typically reported for anthropometric data (often separated by gender)

• Use of these value implicitly assumes a Normal distribution. Assumption is reasonable for most human data.

• Percentiles can easily be calculated from mean and std.dev. using these formulas and/or standard statistical tables (usually z).

Statistics

Percentile

• Commonly used: 5th, 95th, 50th (median)

• Lower-limit dimension: the smaller the system, the more unusable by the largest user Use high percentile

• Upper-limit dimension: the bigger the system, the more unusable by smallest user Use low percentile

Statistics

• Z = (y-)/– Normally distributed with mean = 0 and variance = 1– z is N(0,1)

• From tables of normal cumulative probabilities– P{z≤z(A)} = A

– Example: if zA = 2, A = 0.9772 (two std.dev. above mean is the 97.7%-ile)

– Properties of z:• zA > 0; above mean (>50%-ile)

• zA = 0; at mean (50%-ile)

• zA < 0; below mean (<50%-ile)

Statistics - Standard Normal Variate

Normal Distribution Table

• For female stature (from Table)– = 160.5 cm– = 6.6 cm

• What female stature represents the 37.5th %-ile?– From normal distribution:

z(37.5%) = -0.32

Thus, X(37.5%) = + z = 160.5 - (0.32)(6.6)

= 158.4 cm

Percentile Example

• To combine anthropometric dimension, need to calculate a new distribution for the combined measures, accounting also for the covariance (Cov) between measures (M = mean; S = std. dev.):

Anthropometric Data: Variances

Means add, variances do not!

MX+Y = MX + MY

SX+Y = [SX2 + SY

2 + 2Cov(X,Y)]1/2

SX+Y = [SX2 + SY

2 + 2(rXY)(SX)(SY)]1/2

MX-Y = MX - MY

SX-Y = [SX2 + SY

2 - 2Cov(X,Y)]1/2

SX-Y = [SX2 + SY

2 - 2(rXY)(SX)(SY)]1/2

Class Activity

1. Determine dimensions of product which are critical for design (considering effectiveness, safety and comfort)

2. Determine the related body dimensions3. Select user population (who will use the product or

workplace)4. Conduct reference study to find secondary data, if

available (considering population characteristics) or conduct measurement

5. Select percentile

Anthropometrical Design Procedures

Anthropometric data for individuals is often estimated using stature or body weight in linear regression equations.

Ex: average link lengths as a proportion of body statureAdvantages:

◦ Simplicity

Disadvantages:◦ relationships are not necessarily linear, nor the same for all

individuals◦ Values represent averages for a portion of a specific population

The “Average Human”

• Anthropometric data is most often used to specify reach and clearance dimensions.

• The criterion values most often used:

– Reach: 5% Female

– Clearances: 95% Male

• Try to accommodate as large as possible user population within constraints

Anthropometry in Design

Design for extremes◦ emphasize one 'tail' of distribution

Design for average◦ emphasize the center of a population distribution

Design for adjustability◦ emphasize that all potential users/consumers are

'equal’Varying ranges of accommodation:

◦ 5th-95th %ile: typical◦ 25th-75 %ile: less critical functions or infrequent use◦ 1st - 99th %ile: more critical functions +/- low $◦ 0.01 - 99.99 %ile: risk of severe outcomes

Design Approaches

Example: Door HeightAssuming a normal distribution

◦ z = (X - )/ ◦ Obtain z => %-ile from stats table

What height to accommodate? (95th%-ile male)◦ = 69”; = 2.8” (from anthropometric table)◦ z0.95 = 1.645 = (X - 69)/2.8 => X = 73.6”◦ Additional allowances?

Hair Hats and shoes Gait Etc.

Design for Extremes

leg clearance at a work table

finger clearance for a recessed button

height of an overhead conveyor system

grip size for a power tool

weight of a power tool

height of a conveyor

strength required to turn off an emergency valve

ExamplesWhich design strategy should be employed?

Design for Average:◦ Usually the worst approach: both larger and smaller users

won’t be accommodated

Design for Extremes:◦ Clearance: use 95th percentile male◦ Reach: use 5th percentile female◦ Safety: accommodate >99% of population

Design for Adjustability◦ Preferred method, but range and degrees of adjustment are

difficult to specify

General Strategies and Recommendations

• Working in groups:– Select a workplace near campus. Identify any

‘ergonomic mismatch’. Suggest how the workplace can be better designed from the perspective of engineering anthropometry. You should outline the design approach.

– Pick a journal paper that discusses the use of anthropometric data in design. Submit a one-page summary (in Indonesian) of the paper. Also submit softcopy of the paper.

Homework