17	Environmental Ergonomics

There are numerous factors that can make up a working environment. These include noise, vibration, light, heat and cold, particulates in the air, gases, air pressures, gravity, etc. The applied ergonomist must consider how these factors, in the integrated environment, will affect the human occupants.

K. C. Parsons
2000

INTRODUCTION

So far, we have talked about those issues in workspace design, controls, and tools that have an obvious and direct influence on human performance. But a person’s performance is also influenced by the physical environment in which he or she must carry out a task. Anyone who has tried to mow a lawn in the heat of a summer afternoon or balance a checkbook while a baby is crying can appreciate the influence of these sometimes subtle environmental variables. Often, the effect that the environment will have is not obvious during the design of a workspace or task. Some physical factors will become evident only when the workspace is implemented within the larger work environment.

The study of human factors issues with respect to the physical environment is called environmental ergonomics. According to Hedge (2006), “Environmental ergonomics studies our physiological and behavioral reactions to the ambient environment, and the design of effective barriers that allow us to survive in otherwise inhospitable settings” (p. 1770). By anticipating possible problem areas, such as glare on a visual display screen, human factors experts seek to design tasks and workspaces so that the consequences of noxious environmental variables are minimal. However, despite all attempts to reduce the impact of harmful environmental factors, some issues may arise only through the synergy of the workplace. Action often must be taken by the designer “on the spot” to remedy problems as they are detected. In addition to environmental ergonomics’ focus on the person, “green ergonomics” attempts to remedy environmental problems while simultaneously considering concerns about sustainability of the physical environment (Pilczuk & Barefield, 2014).

In this chapter, we will examine four powerful environmental factors: lighting, noise, vibration, and climate. We encounter these factors within larger environments, such as offices, buildings, and other contained environments. We must also recognize that these factors can be sources of psychological stress, and so they may have harmful physiological and psychological consequences.

LIGHTING

Lighting affects how well people can perform tasks by how it restricts visual perception. However, poor lighting may also be responsible for certain health problems and adverse effects on mood. Human factors experts are most often concerned with the conditions that promote good interior lighting, which is essential in the home and at work. There are some situations where we also must consider lighting conditions outdoors, such as along roadways, for fields and stadiums where sports are played, and so on. In this chapter, we will focus mainly on interior lighting problems.

Lighting considerations are determined by four major human factors issues (Megaw & Bellamy, 1983): (1) how important light levels are to a person’s ability to perform the task; (2) the speed and accuracy with which a person must perform the task; (3) the person’s comfort; and (4) the person’s subjective impressions of the quality of the lighting. As in all design problems, different lighting solutions will be more or less expensive. The human factors expert will need to balance costs against outcomes. “Good” lighting solutions will provide the best visual conditions for the lowest cost.

Our discussion of lighting will cover four topics. First, we will describe how light is measured. Second, we will discuss the characteristics of different kinds of artificial lights. Third, we will talk about how lighting can influence a person’s ability to perform certain tasks. Finally, we will expand on the relationship between lighting and performance in a discussion of the effects of glare, which is the reflection of light from surfaces in the work environment.

LIGHT MEASUREMENT

An evaluation of lighting conditions in a home or work environment must begin with the measurement of effective light intensity, or photometry (Kitsinelis, 2015). However, we must distinguish between light that is reflected from and light that is generated by a surface on which the measurement is made. Illuminance is the amount of light falling on a surface, and luminance is the amount of the light generated by a surface (either a light source or a reflection). Both luminance and illuminance are determined by luminous flux, which is measured in units called lumens. Lumens represent the amount of visible light in a light source and thus the power of the light source corrected for the spectral sensitivity of the visual system. Illuminance is the amount of luminous flux per unit area (one square meter), and luminance is the luminous flux emitted from a light source in a given direction. The luminance of a reflection is a function of both the illuminance and the reflectance of a surface.

Both luminance and illuminance are measured with a device called a photometer. The photometer measures light in the same way that the human visual system does in daylight viewing conditions: each wavelength coming into the photometer is weighted by the corresponding threshold on the spectral sensitivity curve. For measures of luminance, a lens with a small aperture is connected to the photometer. The lens is focused on the surface of interest from any distance. If the light energy within the focused region is not uniform, the photometer integrates over the focused area to give an average luminance. The photometer gives the measure of luminance in candelas per square meter (cd/m²). (A candela is a fixed amount of luminous flux within a fixed cone of measurement.)

For measures of illuminance, an illuminance probe is connected to the photometer and placed on the illuminated surface to be measured, or a special illumination meter can be used. Unlike luminance, the amount of illuminance will vary with the distance of the surface from the light source. The photometer or illumination meter gives the measure of illuminance in lumens per square meter (lm/m²) or lux (lx).

We talked about how important contrast is in Chapter 6. Recall that contrast is the difference in luminance between two areas in the visual field. Contrast can also be measured with the help of a photometer (Kitsinelis, 2015). The contrast (C) between the luminance (L _o) of an object and that of its background (L _b) is often defined as:

$$C=\frac{L_o–L_b}{L_b}$$.

One way of thinking about the difference between luminance and illuminance is that while luminance measures the amount of light coming from a surface, illumination measures the amount of light falling on it. Designers of workspace and workspace lighting are most often concerned with illumination, because it is a measure of the effective amount of light energy for a particular work surface or area. The illuminance for an office should be between 300 and 500 lux at work surfaces, with lower lighting levels needed in homes.

LIGHT SOURCES

Different kinds of light sources will have different illuminances and different costs. One important factor the designer must consider with different light sources is that the accuracy of color perception (which we will call color rendering) will also depend on the light source. Optimal lighting solutions will provide a quality of light that is appropriate for the tasks performed in the environment while minimizing the expense of the lighting system. This means that there will never be a “one-size-fits-all” solution for different work environments. Each environment will potentially require different kinds of lighting.

Daylighting

The most basic distinction we can make is between natural lighting (sunlight) and artificial lighting. Sunlight contains energy across the entire spectrum of visible wavelengths, but with relatively more energy devoted to the long (red) wavelengths. This fact accounts for sunlight’s yellow color. Windows and skylights provide natural lighting in building interiors. Sunlight is inexpensive and allows good color rendering. However, it is not very reliable. Illumination levels will vary as a function of such factors as time of day and year, and the weather. The distribution of natural light cannot be easily controlled. Some workspaces may be easy to position near a window or skylight, but others may not.

Using natural lighting for building interiors is sometimes called daylighting. While natural light through windows and skylights is inexpensive, the heating and cooling costs of daylighting are not negligible. The ratio of skylight to floor area that minimizes annual total building energy use in commercial buildings is estimated to be 0.2 (Nemri & Kwartri, 2006). This means that a 10 × 10 ft. room (100 ft²) should use a skylight no larger than 4 × 5 ft. (20 ft²). While this size of skylight may be appropriate in some circumstances, it may be too small in others, and the problem of light distribution within the room may remain.

There are several architectural options for daylighting. Roof monitors are boxes placed on building roofs through which daylight can enter, and a series of diffusers and mirrors then distribute the light through the building. Light shelves are reflective, horizontal shelves placed above window exteriors that can “catch” and distribute light more evenly through the interior space. Tubular skylights, like roof monitors, use roof-mounted light collectors. The light is then directed down a tube and through a diffusor lens that distributes the light evenly across an interior space.

Most office buildings were constructed without plans for daylighting, and so the light distribution through windows and skylights is not always uniform throughout the interior. One way to remedy this problem is the PSALI (Permanent Supplementary Artificial Lighting Installation) approach. This design approach first analyzes the availability of natural light throughout the interior, with the goal of relying on it as much as possible. Then, artificial lighting is added to supplement the natural light, creating a uniform light distribution over all areas (Hopkinson & Longmore, 1959). With the PSALI approach, more light fixtures would be installed to illuminate desks and areas in the interior that are located farthest from windows, and fewer in the areas that are closest to windows.

Artificial Lighting

Artificial lighting systems are discriminated first and foremost by the kinds of bulbs that the light fixtures use. The two most common types of artificial light bulbs are incandescent and fluorescent (Mumford, 2002).

Incandescent light bulbs were the primary source of home lighting for many years. Their use has been restricted in some countries and they are being phased out in others, including the U.S., because of their energy inefficiency. That is, the number of lumens these bulbs can produce per watt is very low; as much as 95% of the energy that goes into powering the bulb is converted into heat. Because incandescent bulbs that meet more stringent energy efficiency standards are still being manufactured and sold, and because people tend to prefer the qualities of incandescent lighting to those of the alternatives, design issues surrounding incandescent lighting are still relevant in modern home and work environments.

Incandescent light is produced by current flowing through a tungsten filament inside a glass bulb that contains an inert gas, which becomes heated and glows. Tungsten halogen lamps are also incandescent. They differ from ordinary light bulbs in that a halogen gas instead of an inert gas is used in the bulb. Tungsten lamps are long-lasting sources of bright light that are used for vehicle headlights and floodlights (Kitsinelis, 2015).

Incandescent lamps have several benefits, including that their initial cost is low, they generate light energy across the visual spectrum, and they provide a full output of light immediately upon being turned on (Wolska, 2006b). Consequently, despite its low efficiency, incandescent lighting has been useful for residences, where only a small number of lamps are needed, but not for businesses and other larger organizations that may have much greater lighting demands.

Fluorescent lamps are gas discharge bulbs. An electric current is alternated through an inert gas, producing invisible ultraviolet light, which excites phosphors coating the inside of the bulb. Although the light appears steady, it actually flickers at a high frequency (equivalent to the frequency of the alternating current). Fluorescent lamps can require as little as 25% of the power required to light an incandescent lamp, and they have a longer life. Because of their relative efficiency, fluorescent lamps are used for lighting in schools, offices, and industrial buildings, and bulb-like compact fluorescent lamps are now replacing incandescent bulbs for home use.

One disadvantage of fluorescent lamps is that their light output decreases over the lamp’s lifespan (Wolska, 2006b). Also, the spectral distribution for the common cool-white fluorescent lamp is not much like that of daylight, which results in poor color rendering. Mayr, Köpper, and Buchner (2014) showed that people can’t discriminate between colors as well under compact fluorescent lighting as they can under halogen lamps. An alternative to cool-white fluorescent bulbs are more expensive full-spectrum or high color rendition bulbs, which use a mix of different phosphors to more closely match the spectrum of natural light. Fluorescent lighting technology has continuously improved, leading Mumford to state as far back as 2002 that, “Compared with the fluorescent lighting available only 20 years ago, there are fluorescent products available today which reduce office and video terminal fatigue, are economical in use and can help some of the readers with special needs” (p. 3). You should be aware that there are more speculative claims about the benefits of full-spectrum fluorescent lighting for a person’s performance of perceptual and cognitive tasks, and even a person’s physical and psychological health! However, there is little scientific evidence for these purported benefits (McColl & Veitch, 2001; Veitch & McColl, 2001).

Although they are more efficient than incandescent lamps, fluorescent lamps are still not very efficient. There are other gas discharge lamps, including induction, metal halide, and sodium lamps, that are far more efficient. However, these other lamps are much more expensive, and their color rendering capabilities may be much poorer because of peaks in their spectral frequency distributions. Figure 17.1 shows the spectra for four types of lamps: incandescent, fluorescent, metal halide, and sodium. The incandescent and fluorescent spectra are much smoother than those of the metal halide and sodium lamps. The extreme peaks in the metal halide and sodium spectra mean that these lamps provide large amounts of light from only a few wavelengths. These few wavelengths can “wash” the work area with the hues of those wavelengths, making accurate color rendering difficult or impossible. Low-pressure sodium lamps, for example, are used almost exclusively for street lights. They generate yellow/orange light and provide no color rendering (Wolska, 2006b).

FIGURE 17.1Light spectra for incandescent, fluorescent, metal halide, and sodium lamps.

The last kind of lighting we will discuss is called solid-state lighting, and it uses arrays of light emitting diodes (LEDs). LED bulbs can fit into the sockets of light fixtures manufactured for incandescent bulbs. LEDs are particularly useful for applications where small size and long lifetimes are important, such as for color indicator lamps. For example, they are now widely used for traffic lights, toll booth lane indicators, light rail signals, vehicle tail lights, and airport runway lighting (Boyce, 2014). While high-brightness LEDs can be used for a variety of lighting applications, they have a number of drawbacks (Žukauskas et al., 2002). In addition to the fact that they can be very expensive, like fluorescent bulbs, color rendering can be poor, and there is some evidence that blue and cool-white LEDs can cause glare, damage the retina, and interfere with normal sleep cycles (Algvere, Marshall, & Seregard, 2006; American Medical Association, 2016).

Apart from the kind of bulb (incandescent, fluorescent, gas discharge, or solid-state) that a light source uses, the next common distinction between light sources is whether the lighting is direct or indirect (Wolska, 2006b). Direct lighting falls on a surface directly from the light source. In contrast, indirect lighting has been reflected from other surfaces, often the ceiling, before falling on the work surface. Technically, if 90% or more of the light from a source is directed toward the work surface (downward), then the lighting is classified as direct. If 90% of the light is directed away from the work surface (upward), then the lighting is classified as indirect. Lighting is called semi-direct or semi-indirect when 60%–90% of the light is directed toward or away from the work surface, respectively. Direct lighting often results in glare, and indirect lighting is not very effective for work requiring close visual inspection.

ILLUMINATION AND PERFORMANCE

We have emphasized throughout this book that the tasks that people perform have perceptual, cognitive, and motor information-processing components. We have also discussed how tasks can be classified according to the extent to which each of these components is required for a task’s completion. Illumination will have the greatest effect on those tasks that depend on the visual perception component.

We can characterize the visual difficulty of a task by the size of the smallest critical details or items with which a person must work and the contrast of those details with the background. Often, simply increasing illumination does not make a visually difficult task easy. Figure 17.2 shows the effects of object size, illuminance, and contrast on performance (Weston, 1945). While performance improves as size, illumination, and contrast increase, performance with small, low-contrast objects is always much poorer than that for larger, higher-contrast objects.

FIGURE 17.2The effects of size (in minutes of visual angle), illuminance, and contrast on performance. Black circles = high contrast; triangles = medium contrast; gray circles = low contrast.

Many kinds of field studies are conducted to directly evaluate performance under different levels of illumination and types of lighting. For example, one study measured productivity under changes of lighting in a leather factory over a 4-year period (Stenzel, 1962). The workers’ tasks involved cutting shapes from skins to make leather goods such as purses and wallets. For the first 2 years of the study, work was performed in daylight with additional fluorescent fixtures, which provided an overall illuminance of 350 lx on the factory floor. Before the third year, the daylight was removed, and fluorescent lighting was installed that provided a uniform 1000-lx illumination. As shown in Figure 17.3, workers’ performance clearly improved after the installation of the lighting.

FIGURE 17.3Normalized performance in a leather factory as a function of lighting conditions (left panel is old lighting, right panel is new lighting) and month (from July through June).

Can we conclude that the increase in illumination level caused the increase in productivity? Unfortunately not, although the findings suggest that this was the case. The change in performance could be due to other factors, such as the increased uniformity of illumination, color modifications, or unrelated variables (such as pay raises or different work schedules) that may have been altered at about the same time as the change in lighting.

The well-known Hawthorne lighting experiments demonstrated how hard it is to control extraneous variables in the work environment. Three experiments were conducted from 1924 to 1927 at the Hawthorne Plant of the Western Electric Company to assess the effects of lighting on productivity (Gillespie, 1991). The impetus for the experiments came from the electrical industry, which claimed that good lighting would increase productivity significantly. The workers were informed about the nature of the study in order to obtain their cooperation.

In the initial experiment, three test groups of workers involved in the assembly of telephone parts performed under higher than normal lighting levels, while a control group performed under normal lighting. Production increased dramatically in the three test groups in comparison to their level of productivity before the experiment, but it increased a similar amount for the control group. Also, within each experimental group, there was no correlation between productivity and the lighting levels under which each group performed. The researchers concluded that the increased productivity was due to an increase in management’s involvement with the workers, which was required for measuring lighting levels and productivity, rather than to the lighting level itself. That is, because the workers either knew they were being more closely watched by their managers or were enjoying the increased attention being given to them by their managers, they worked harder than they had before the lighting experiment began.

The researchers in charge of the Hawthorne study conducted more experiments, in which they made explicit attempts to control the effect of management attention. Even in these experiments, the lighting level had little effect on performance, except at very low illumination levels. One explanation for this is that the workers may have expended more effort than usual under conditions of low illumination to compensate for any increased difficulty.

There are several other mechanisms by which new lighting may affect performance in the work environment (Juslén & Tenner, 2005). These include visual comfort, visual ambience, interpersonal relationships, and job satisfaction, as well as biological effects such as the timing of circadian rhythms and alertness (Boyce, 2014; van Bommel, 2006; see also Chapter 18). One study showed that the productivity of assembly workers increased when their work stations were equipped with a controllable task-lighting system, which allowed the workers to adjust the lighting to their preferred intensity levels (Juslén, Wouters, & Tenner, 2007). The increase in productivity could have been due to improved visual perception or to some other psychological or biological mechanism associated with the controllable lighting system. Juslén et al. argue: “Seeing lighting change as a process with several mechanisms, which are partly ‘light related mechanisms’ and partly general mechanisms, will help designers and managers to estimate whether a lighting change is worth the investment” (p. 853).

While it is difficult to draw conclusions about the relationship between lighting and task performance in field studies, we can be much more confident about what we observe when we move the task environment into the laboratory. This also requires that the researchers design a simulation of the real-world task that preserves its critical elements while eliminating those that would make a causal relationship difficult to establish. Stenzel and Sommer (1969) performed such a laboratory experiment, in which people either sorted screws or crocheted stoles. They varied illuminance during task performance from 100 to 1700 lx. The number of errors that were observed decreased with increasing illuminance for crocheting but not for sorting. For the sorting task, errors decreased as illuminance increased to 700 lx, but then increased as illuminance rose to 1700 lx. Therefore, the effect of increasing illumination depended on the specific task that was performed.

Illumination and contrast are particularly critical in designing workplaces appropriate for older adults, because visual acuity declines rapidly with age. One study asked young (18–22 years) and older (49–62 years) adults to proofread paragraphs for misspelled words (Smith & Rea, 1978). The researchers presented paragraphs that were of good-, fair-, or poor-quality text on white or blue paper, and varied illumination from 10 to 4900 lx. The readers’ performance increased as the copy quality increased and also as the illumination increased. However, the young adults showed very little improvement with increased illumination, whereas the older adults showed marked improvement (see Figure 17.4). Therefore, illumination and print quality are more important for older than for younger people. This fact has been confirmed in other studies showing that a higher level of illumination is more critical for older adults who are visually impaired through age-related macular degeneration than for those who have normal vision for their age (Fosse & Valberg, 2004).

FIGURE 17.4Proofreading performance by younger and older adults as a function of illumination level.

Performance is not the only issue involved in designing a lighting scheme. Visual “comfort” is also important. A visual environment is comfortable when workers in that environment can complete perceptual tasks with little effort or distraction, and without stress; that is, when sources of visual discomfort are absent. Visual discomfort can occur when the visual task is difficult (e.g., having to resolve fine details; driving under foggy conditions), irrelevant objects provide distraction by attracting the worker’s attention away from the task, and lighting conditions produce confusable reflections on the objects in the workspace (Boyce, 2014).

An important factor in predicting visual comfort is the ratio of the luminance of an object or task area being viewed to the luminance of its surroundings. Visual comfort can be maintained as long as the luminance ratio does not exceed 5:1 (Cushman & Crist, 1987). However, comfort and performance may not be significantly affected even when the luminance ratio is as large as 110:1. Cushman, Dunn, and Peters (1984) had people make prints of photograph negatives under luminance ratios ranging from 3.4:1 to 110:1. As the luminance ratio decreased, the printing rate declined slightly, but so did the error rate. The study participants reported less ocular discomfort and overall fatigue when they were allowed to adjust the luminance ratio.

Much of what makes one lighting scheme preferable to another is subjective: that is, it cannot be objectively measured by a designer. Some environmental qualities, such as clarity, pleasantness, spaciousness, and how relaxing a space is, are not functions of luminance flux. Flynn (1977) published the results from studies in which he asked people to give subjective ratings of these and other qualities under different types of lighting. The lighting schemes used in these studies varied along several dimensions. An overhead versus peripheral dimension determined whether the lights were mounted on the ceiling or on the wall. A nonuniform versus uniform dimension described the distribution of light in the room as a function of the location of objects and surfaces in the office. Lighting was also adjusted to be either bright or dim, and either warm or cool. Table 17.1 shows how the values of different lighting dimensions evoke positive qualities of, for example, spaciousness and privacy. Some of these qualities will be more important than others, depending on the task.

TABLE 17.1

Lighting Reinforcement of Subjective Effects

Subjective Impression	Reinforcing Lighting Modes
Visual clarity	Bright, uniform lighting mode
	Some peripheral emphasis, such as with high-reflectance walls or wall lighting
Spaciousness	Uniform, peripheral (wall) lighting
	Brightness is a reinforcing factor, but not a decisive one
Relaxation	Nonuniform lighting mode
	Peripheral (wall) emphasis, rather than overhead lighting
Privacy or intimacy	Nonuniform lighting mode
	Tendency toward low light intensities in the immediate locale of the user, with higher brightnesses remote from the user
	Peripheral (wall) emphasis is a reinforcing factor, but not a decisive one
Pleasantness and preference	Nonuniform lighting mode
	Peripheral (wall) emphasis

Hedge, Sims, and Becker (1995) conducted a field study investigating productivity and comfort with two different lighting systems installed in a large, windowless office building. These were lensed-indirect uplighting (LIL) and direct parabolic lighting (DPL). The LIL used fixtures suspended from the ceiling, which projected light upward to be reflected from the walls and ceiling. The DPL used fixtures recessed into the ceiling and shielded by parabolic louvers. Office workers responded to a questionnaire that asked them about their satisfaction with the lighting system installed in their offices. The DPL system generated significantly more complaints than the LIL system about problems like glare and harshness, and workers estimated up to four times more productivity loss because of such lighting problems. Workers in the DPL group also reported three to four times more productivity loss due to visual health problems, such as focusing problems, watery eyes, or tiredness.

GLARE

Glare is high-intensity light that can cause discomfort and interfere with the perception of objects of lower intensity. There are different kinds of glare: direct and reflected. Light sources within the visual field, such as windows and light fixtures, can produce direct glare. Reflected glare is produced by objects and surfaces that reflect light. Reflected glare can be avoided by locating light sources and work surfaces so that light sources are not in an “offending zone.” The offending zone is where light from the source will reflect from the work surface into the eyes (see Figure 17.5).

FIGURE 17.5The offending zone for glare.

There are different kinds of reflected glare. Specular reflection produces images of objects in the room on the viewing surface. Veiling reflection results in a complete reduction of contrast over parts of the viewed surface. Both direct and reflected glare can be particularly debilitating for workstations with visual display units (VDUs).

Glare also can be classified according to its severity. Disability glare reduces the contrast ratio of display characters by increasing the luminance of both the display background and the characters. This reduces the detectability, legibility, and readability of the display characters. It usually results from a light source that is located close to the line of sight. Discomfort glare, which may or may not be accompanied by disability glare, will cause the worker discomfort when the work surface is viewed for a period of time.

Discomfort from glare increases as the luminance and number of glare sources increase (Wolska, 2006a). However, because discomfort is a subjective event, it will be affected by many factors other than light intensity. For instance, a person will report greater discomfort produced by glare when she is performing a visually demanding task than when the task is not as visually demanding. Her prior experience with the task may be important as well.

For an example illustrating the role of prior experience, consider the fact that automobiles in Europe have low-intensity amber (filtered) headlights. U.S. automobiles, in comparison, have very bright white (unfiltered) headlights. Sivak, Olson, and Zeltner (1989) reasoned that European drivers, because of their experience with low-intensity amber headlights, may be more subject to discomfort glare than U.S. drivers when driving on U.S. roadways. This was the case; West German drivers rated filtered and unfiltered headlights of different brightnesses higher in discomfort than did U.S. drivers. The drivers’ past experience helped determine the degree of discomfort.

There are several measures of visual discomfort. One that can be used to assess the potential for direct discomfort glare is the visual comfort probability (VCP) method (Guth, 1963). These measures take into account the direction, luminance, and solid angle of the glare source, as well as the background luminance. The VCP method relies on calculation of a glare sensation index (M):

$$M=\frac{L_{S}Q}{\text{2}P{F}^{0.\text{44}}}$$,

where:
LS	is the luminance of the glare source,
P	is an index of the position of the glare source from the line of sight,
F	is the luminance of the entire field of view including the glare source;

Q=20.4 ω_s +1.52ω_s^0.2 − 0.075,

and

where Ω_S is the (solid) visual angle of the glare source. Sometimes, there may be more than one glare source affecting a single location. In this case, the glare sensation M _i for each source (i = 1, 2, 3,..., n) at that location can be calculated and the results compounded into the single discomfort glare rating (DGR) by the formula

$$DGR={\left[{\displaystyle \sum _{i=1}^n\left(M_i\right)}\right]}^a,$$

where n is the number of glare sources

$$a=n^{-0.0914}.$$

and

The VCP is defined as the percentage of people who would find the level of direct glare in the environment acceptable. The DGR can be converted into VCP using the formula

$$VCP=\text{279}-\text{11}0({\text{log}}_{\text{1}0}DGR)$$

for the range of primary concern VCP from 20% to 80%. It is generally agreed that direct glare will not be a problem for a lighting application if the VCP is 70 or higher.

We can reduce glare in many ways. Window luminance can be controlled with blinds or shades. Similarly, shades and baffles on light fixtures can reduce the amount of light coming directly from the fixtures. We can position VDUs or other displays so that bright sources of light are not in the field of view and reflections are not seen on the screen. Some displays allow the user to avoid glare by tilting or swiveling the screen. Anti-glare devices, such as screen filters, can be used for VDUs, but they reduce contrast and so may degrade visibility. Liquid crystal displays (LCDs) can replace older cathode ray tube (CRT)-type displays and are less susceptible to glare.

Glare is a significant factor in night driving. Direct glare from the headlights of oncoming vehicles and indirect glare from the headlights of trailing vehicles can produce discomfort and impair a driver’s vision. A study examined how well drivers performed with direct glare by mounting a simulated headlight source on the hood of an instrumented vehicle (Theeuwes, Alferdinck, & Perel, 2002). In direct glare conditions, when the simulated headlight was turned on, drivers drove more slowly and were less likely to detect pedestrians at the side of the road. Older drivers were more adversely affected than younger ones. Another study showed that older adults reported more discomfort from the same levels of glare in driver-side mirrors than did younger adults. When the flat mirrors were replaced with curved ones, both older and younger adults suffered less discomfort (Lockhart, Atsumi, Ghosh, Mekaroonreung, & Spaulding, 2006).

Another study examined how well truck drivers performed with indirect glare reflected in the side-mounted rearview mirrors of a truck simulator (Ranney, Simmons, & Masalonis, 2000). Drivers were asked to detect stationary pedestrians and to determine the location of a target X presented on vehicles in the truck’s mirrors. The researchers created glare by directing beams of light into the side mirrors, which were set for either no glare reduction or high glare reduction (which reduced the reflectiveness by 80%). When the mirrors did not reduce glare, truckers could not detect targets in the mirror well, and they had poorer control of the truck: their lane variability increased, speed on curves slowed, and steering variability increased. However, glare-reducing mirrors did not improve the truckers’ performance much, either in target detection or in vehicular control. Nonetheless, the drivers indicated that they preferred having the glare-reducing mirrors.

NOISE

Noise is undesirable background sound that is irrelevant to the task that someone is trying to accomplish (School, 2006). It is present to some extent in any work environment, as well as in almost all other settings. Noise can be generated by office equipment, machinery, conversation, and ventilation systems, as well as by traffic and miscellaneous events such as doors slamming. A high noise level can be uncomfortable; it can also reduce performance, and workers may experience permanent hearing loss. Human factors experts can help design and modify work environments to reduce the deleterious effects of noise by determining what noise is tolerable and, as with lighting, establish suitable aesthetic criteria for the well-being of the people who work in the environment.

In this section, we will first discuss how noise levels are measured, and then how noise can affect a person’s performance. Then we will discuss how noise causes hearing loss, and the effects of hearing loss on performance. Finally, we will discuss strategies for reducing noise in the workplace.

NOISE MEASUREMENT

Remember that an auditory stimulus (a tone or a sound) can be broken into its component frequencies just as a light source can be broken into its component wavelengths. Each frequency in a sound will have an amplitude that describes how much of that frequency contributes to the sound as a whole. When we measure the intensity (amplitude) of a noise, we have to worry about these different frequencies, because people are better at hearing some frequencies than others.

A sound-level meter (see Figure 17.6) will give a single measure of sound amplitude averaged over the auditory spectrum. Just as the photometer is calibrated for human sensitivity to light of different wavelengths, the sound-level meter is calibrated according to human sensitivity to tones of different frequencies. However, remember also that relative sensitivity (the loudness a person perceives for tones of different frequencies) is a function of the amplitude of a tone. This means that the sound-level meter will need to be calibrated differently to measure noise at different intensity levels.

FIGURE 17.6Model CEL-354 Sound-Level Meter.

A sound-level meter often has three calibration scales, one appropriate for low intensities (the A scale), one for intermediate intensities (the B scale), and one for high intensities (the C scale), although the B scale is omitted from some meters. Figure 17.7 shows how the scales differ, and that the difference is primarily in how the meter weights frequencies below 500 Hz. If we measure the same sound twice, using the A and the C scales, the difference between the two measurements gives an indication of the intensity of low-frequency components in the sound. If the two measures are very similar, then the sound energy contains components that are mostly above 500 Hz, whereas if the C measure is much higher than the A measure, then a substantial portion of the sound energy is below 500 Hz. Some sophisticated meters include band-pass filters that let us measure sound energy within specified frequency regions.

FIGURE 17.7The spectral weightings for the A, B, and C scales on a sound-level meter.

In most environments, noise levels will not be constant but will fluctuate, either quite rapidly or more gradually over time. Most sound-level meters can accommodate changes like this because they are equipped with “slow” and “fast” settings that differ in the length of the time interval over which the noise is averaged. The meter will average the noise for a longer period of time on the slow setting (1 s) than the fast setting (125 ms). If the noise level changes rapidly, the slow setting will show less fluctuation on the meter. Some meters have “hold” buttons to use with the fast setting that will display maximum and/or minimum intensity levels.

We might also be concerned about a person’s total noise exposure across the course of a day. We can get a cumulative measure of a worker’s total exposure with a device called an audio-dosimeter (Casali, 2006), which is worn by the worker for an entire day. These meters are small and inexpensive, but their measurements can be inaccurate. This is because the meter will register high noise levels that arise because the noise source is close to the microphone. Although some of these sounds may be of concern, others, such as the worker’s own voice, may not be.

NOISE LEVEL AND PERFORMANCE

A person’s performance may suffer in many ways if he is forced to work in a very noisy environment. We discussed in Chapter 8 how noise can mask both speech and nonspeech sounds. Masking will interfere with a person’s attempts to communicate with other people and to perceive auditory displays. When a person shouts to try to overcome a high background noise level, his speech patterns will change, and these changes will also impair communication. Even when a worker is not trying to communicate with anyone else, other people’s conversation in the background can prevent him from concentrating on reading or listening to other verbal material.

Noise can evoke highly emotional responses. Anyone who has been exposed to the sound of fingernails being scraped over a blackboard can appreciate how compelling some sounds can be. The startle reaction, for example, is something that everyone experiences when they hear a sudden loud noise. It consists of muscle contractions and changes in heart and respiration rate, and is usually followed by an increase in arousal. Fortunately, such reactions are usually very brief, and their intensity tends to diminish with repeated exposure.

Sonic booms are examples of unpleasant noise that evoke strong emotional responses. A sonic boom occurs when an aircraft travels faster than the speed of sound. The booms occur unexpectedly, have rapid onset, and are loud enough to shake buildings and startle people. One of the most notorious studies of the effects of sonic booms on people was conducted in 1964 by the U.S. Federal Aviation Administration (FAA; Borsky, 1965). From February 3 through July 30 of that year, residents of Oklahoma City, which during the latter part included one of the authors (RWP), were subjected to eight sonic booms per day to assess the possible effects of supersonic transport flights on residents’ attitudes. Interviews were conducted with nearly 3000 persons at 11, 19, and 25 weeks after the beginning of the testing period, and complaints filed by all residents were recorded. As Gordon Bain, then Deputy Administrator for Supersonic Transport Development of the FAA, commented, “The Oklahoma City sonic boom study … was the first major effort anywhere in the world to determine the nature of public reaction to sonic boom at specified, measured levels over a reasonably extended period of time” (Borsky, 1965, p. ix).

Almost all of the respondents reported that the booms rattled and shook their houses, and the booms broke many windows in the city’s largest buildings. Otherwise, physical damage was minimal. Some 35% of the interviewees reported having startle and fear responses to the sonic booms, and 10%–15% reported interference with communication, rest, and sleep. Only 37% indicated annoyance with the booms during the first interview period, but by the last period, more than half (56%) did. This suggests that sonic booms may in fact become more annoying with prolonged exposure. But, because the intensity of the booms was increased after each interview, the increased annoyance could have been due to the increase in boom intensity.

At the last interview, approximately 75% of the residents indicated that they did not find the eight booms per day too bothersome, but 25% said that they did. Moreover, 3% of the entire population, or about 15,000 people, were sufficiently bothered to file a formal complaint or lawsuit. This number is most likely an underestimate, as the report notes that one reason for the low complaint level was that “there was widespread ignorance about where to complain” (Borsky, 1965, p. 2).

Not all emotional responses are necessarily detrimental to performance. Background noise that increases arousal will produce better performance on a vigilance task, in which performance tends to decline as arousal decreases (see Chapter 9; McGrath, 1963). However, this is not true for all vigilance tasks. Some researchers have found that vigilance performance can sometimes be worse with noise (e.g., Becker, Warm, Dember, & Hancock, 1995).

Noise levels as low as 80 dB (about as noisy as a vacuum cleaner) can have a detrimental effect on performance. People may have trouble with the following activities if they try to do them in a noisy environment: (1) tasks of extended duration, if the background noise is continuous; (2) tasks that require a steady gaze or fixed posture, which can be disrupted if a person is startled by sudden noise; (3) unimportant or infrequent tasks; (4) tasks that require comprehension of verbal material; and (5) open tasks, in which a rapid change of response may be required (Jones & Broadbent, 1987).

A comprehensive evaluation of the noise levels in an environment can, therefore, be complicated. This is because the acceptability of different noise levels depends on the task to be performed, and the way that noise levels are measured depends on the background intensity and frequencies of other noise in the environment. These background noises, produced by mechanical systems such as air conditioners, can also generate intense low-frequency sound waves that vibrate floors and walls. These vibrations produce rattles, and even audible noise, called rumble.

There are several established methods for rating noise and assessing its acceptability (Broner, 2005), and each method is based on “noise criterion” curves like the ones shown in Figure 17.8. A noise criterion specifies the maximum intensity level for noise of different frequencies in different environments that will not interfere with speech or be otherwise disturbing.

FIGURE 17.8Balanced noise criterion curves.

The specific noise criterion curves shown in Figure 17.8 were developed by Beranek (1989). They are called the Balanced Noise Criterion (NCB) curves, intended to be applicable to vehicles and buildings. Noise frequencies in a task environment are measured in octave bands, which are ranges of frequencies from one half to double the reference or “center” frequency. So, for example, the frequencies measured for a 500-Hz octave band center frequency range from 250 to 1000 Hz. Each NCB corresponds to a different kind of environment, with louder environments allowing higher intensities before exceeding the NCB values. Similarly, lower noise frequencies can have higher intensities before exceeding any NCB value. The A and B regions to the upper left of the figure indicate those combinations of intensity and frequency that produce clearly and moderately noticeable vibrations, respectively.

To use the curves, we must first decide what the appropriate NCB level is for the environment in question. We do this by consulting tables such as Table 17.2. For example, the environment might be a telemarketing office, and the task people are expected to perform is talking with potential clients on the telephone. We might decide that this not quiet but not loud environment, with many people talking on the telephone, rates an NCB level of 35 (“moderately good listening conditions” in Table 17.2). We will then measure the sound levels in decibels for each octave frequency band. If the average of the noise levels in the four bands most important for speech (the 500-, 1000-, 2000-, and 4000-Hz bands) exceeds the value of the chosen NCB (in this case, 35), then the environment is too noisy, and we will need to take measures to reduce the noise. We evaluate rumble similarly by considering only the sound-pressure levels in the octave bands of 31.5–500 Hz, and vibration by determining whether the levels in any of the three lowest frequency octave bands fall in the A or B regions.

TABLE 17.2

Recommended NCB Curves and Sound-Pressure Levels for Several Categories of Activity

Acoustical Requirements	NCB Curvea	Approximateb LA (dBA)
Listening to faint music or distant microphone pickup used	10–20	21–30
Excellent listening conditions	Not to exceed 20	Not to exceed 30
Close microphone pickup only	Not to exceed 25	Not to exceed 34
Good listening conditions	Not to exceed 35	Not to exceed 42
Sleeping, resting, and relaxing	25–40	34–47
Conversing or listening to radio and TV	30–40	38–47
Moderately good listening conditions	35–45	42–52
Fair listening conditions	40–50	47–56
Moderately fair listening conditions	45–55	52–61
Just acceptable speech and telephone communication	50–60	56–66
Speech not required but no risk of hearing damage	60–75	66–80
aNCB curves are used in many installations for establishing noise spectra. bThese levels (LA) are to be used only for approximate estimates, since the overall sound-pressure level does not give an indication of the spectrum.

The NCB curves are widely used in the U.S., but a similar set of curves, called the Noise Rating curves, proposed by Kosten and Van Os (1962), are used more extensively in Europe. The Noise Rating curves weight the intensity of each octave frequency band to correct for the sensitivity of the auditory system. This system, therefore, gives higher weight to higher frequencies. The Noise Rating system also allows “correction” in the curves depending on the duration (e.g., noise is present 5% of the time) and quality of the noise (e.g., intermittent vs. continuous). There are other noise assessment methods that focus on other dimensions of the acoustic environment. Room Criterion curves, for example, optimize sound quality (Blazier, 1981, 1997).

Which noise assessment method you use will depend on the environment you are evaluating and the priorities of the people working in that environment. An evaluation of the noise levels of a concert hall, for example, will not require you to resolve the same kinds of issues as an evaluation of a factory floor. Whichever method you choose, the first step toward optimal acoustic design of a workplace is the measurement and evaluation of noise levels.

HEARING LOSS

Many popular musicians, including Sting, Pete Townshend, Jeff Beck, Eric Clapton, and will.I.am, have severely impaired hearing and tinnitus (discussed below) from their many years of standing on stage in front of their amplifiers. For these musicians, the constant exposure to very high noise levels resulted in a permanent decrease in their auditory sensitivity. Such decreases are called threshold shifts (Haslegrave, 2005). Short-term exposure to high noise levels can cause temporary threshold shifts in anyone. A temporary threshold shift is defined as an elevation in a person’s auditory threshold measured 2 minutes after exposure. The magnitude of the temporary threshold shift is a function of the noise level and frequency, and the length of exposure time (see Figure 17.9).

FIGURE 17.9Temporary threshold shift as a function of noise level, frequency, and exposure time.

Human factors engineers need to determine whether the noise exposures in the environments they are studying are large enough that short-term hearing impairment is possible. For example, military planes do not usually have insulated cockpits, and so engine and wind noise requires pilots to wear ear protection. Nonetheless, the U.S. Air Force routinely loses many pilots a year because of permanent hearing loss. With earplugs or other suitable ear protection, pilots can avoid permanent damage, but may still experience a temporary threshold shift (Kuronen, Sorri, Paakkonen, & Muhli, 2003). These shifts, which occur for pilots flying many different kinds of aircraft, are small enough that the pilots are not at high risk for permanent threshold shifts.

A permanent threshold shift is an irreversible increase in the auditory threshold; that is, permanent damage. The magnitude of a permanent threshold shift will depend on the years of exposure and the frequencies in the noise. A person’s degree of hearing loss is quantified by the magnitude of the threshold shift, with up to 40 dB impairment considered “mild,” between 41 and 55 dB “moderate,” between 56 and 70 dB “moderately severe,” between 71 and 90 dB “severe,” and 90 dB or greater “profound.” Usually, hearing loss due to long-term noise exposure is concentrated on frequencies around 4000 Hz. Figure 17.10 shows hearing losses in workers in a jute-weaving factory. Workers who had been in the factory the longest (some over 50 years) showed moderate damage for sounds varying from 500 to 6000 Hz, with the most severe losses for frequencies around 4000 Hz.

FIGURE 17.10Permanent threshold shift as a function of exposure duration and noise frequency.

The relationship between noise intensity and frequencies is shown in Figure 17.11. This figure shows permanent threshold shifts for workers exposed to different noise levels and different frequencies for 8-hour shifts over 10 years. The accumulated effect of exposure to 80 dB noise for 10 years of 8-hour shifts is negligible (Passchier-Vermeer, 1974), but these effects increase dramatically for noise levels of 85 dB and above. Figure 17.12 shows the maximum amount of time that a worker can be exposed to potentially damaging noise levels without producing a permanent threshold shift. This maximum exposure duration decreases with increasing decibels.

FIGURE 17.11Accumulated effects of exposure to noise (noise-induced permanent threshold shifts) of different intensity levels for 10 years.

FIGURE 17.12The maximum acceptable exposure duration to high-level noise.

In Chapter 7, we discussed how delicate the anatomy of the inner ear is. We also discussed how sound energy is nothing more or less than changes in air pressure that push against these delicate structures. Sudden loud sounds deliver extremely high pressures to the inner ear, like poking the mechanism with a pencil. Therefore, these kinds of noises may result in acoustic trauma and permanent damage to the structures of the inner ear.

Consider the effect of sounds like gunshots and claxons, for which the onset of the sound is rapid or stepped. Combat soldiers, for example, are frequently exposed to noises like these when they shoot their rifles or when they are in close proximity to bomb blasts. One study reported that of 29% of soldiers exposed to acute acoustic trauma while in the military were still experiencing tinnitus (ringing in the ears) when discharged (Mrena, Savolainen, Kuokkanen, & Ylikoski, 2002). Moreover, of this group, more than 60% reported still experiencing tinnitus 10 years later. A 2005 report from the U.S. Institute of Medicine indicated that 62% of soldiers treated for blast injuries also experience acute acoustic trauma. This report also estimated the number of veterans with permanent damage at over 25%. The consequences of acute acoustic trauma can be severe: Disability payments to U.S. veterans with hearing loss total approximately $1 billion dollars annually, and soldiers experiencing tinnitus describe it as a source of difficulties in their lives (Schutte, 2006).

NOISE REDUCTION

Because the effects of noise can have such profound physical consequences, reduction of noise is a fundamental concern in human factors engineering. Machinery and equipment designers should work to minimize the noise output of their products. After all engineering efforts have been exhausted, workers can be protected from noisy equipment by baffles, which provide a physical sound-absorbing barrier between the worker and the source of the noise, and by ear protection devices.

Ear protection devices are the simplest resource available for noise control. These devices fall into two categories: earplugs and earmuffs. Several types of earplugs and earmuffs are readily available over the counter, and the degree of sound reduction that they are supposed to provide is usually clearly marked on the packaging. However, the level of protection they provide is frequently less than the manufacturers’ ratings (Casali, 2006; Park & Casali, 1991). One solution to this problem is to use earmuffs, which are usually more expensive than earplugs but more effective. Another solution is to use custom-molded earplugs, which, because they can only be inserted correctly, provide better fit and protection than standard earplugs (Bjorn, 2004).

The reason why standard earplugs are not effective as they should be is probably that users do not know how to fit the earplugs properly (Park & Casali, 1991). Noise reduction ratings for three types of earplugs and a popular earmuff with and without user training are shown in Figure 17.13, together with the manufacturers’ ratings. The earmuff provided more protection for untrained users, probably because the earmuff is easy to fit. For users trained on its fit, however, a malleable foam plug, the earplug provided maximum noise reduction. Note, though, that for both trained and untrained users, the noise reduction ratings were uniformly less than those claimed by the manufacturers.

FIGURE 17.13Noise reduction ratings for four hearing protection devices (HPDs) for trained and untrained subjects and as provided by the manufacturer.

Although earmuffs are generally effective at attenuating sound, they are more effective for ­frequencies above 2000 Hz than for those below (Zannin & Gerges, 2006). A benefit of earmuffs is that, when combined with headphones, they can provide both a source of protection against external noise and a means to deliver acoustic information in a noisy environment. Custom-molded earplugs with built-in electronics are also available, but they can be very expensive. Noise-cancelling headphones are equipped with active noise control (Casali, Robinson, Dabney, & Gauger, 2004). For such headphones, a microphone senses the frequency and amplitude of the sound inside the headphones and then produces an inverted signal, 180° out of phase with the sensed one, which cancels out the energy of the sound wave. Active noise reduction works best for repetitive noise and for low-frequency noise below 1000 Hz. Noise-reduction headphones have been shown to reduce miscommunication errors in pilots (Jang, Molesworth, Burgess, & Estival, 2014).

There are two issues that we need to keep in mind when trying to minimize people’s exposure to noise, regardless of the type of ear protection we eventually decide is most appropriate. First, because of the discrepancy between manufacturers’ noise reduction ratings and the attenuation any ear protection device actually provides, we must allow a large safety margin between the actual noise level and the noise reduction rating of the device. Second, the user needs to be trained to use the ear protection device and made to understand why ear protection is important (Behar, Chasi, & Cheesman, 2000; Park & Casali, 1991). Many companies and agencies (included the U.S. Armed Forces) now have hearing conservation programs, in which the use of ear protection devices is only part of an overall training program that emphasizes compliance. In companies that do not have comprehensive programs, 40%–70% of workers typically will not use the devices at all.

VIBRATION

The term vibration refers to any oscillatory motion around a central point and is usually described in three dimensions. As with sound waves, vibration can be characterized in terms of amplitude and frequency. Often, the same mechanisms that produce noise also produce mechanical vibration. If this vibration is in a piece of equipment or machinery, the operator will also experience vibration. For example, offshore oil rigs vibrate a lot and are noisy, and so all the workers stationed on a rig are exposed to quite a bit of noise and vibration (Health & Safety Executive, 2002).

Vibration is measured with an accelerometer, which can be attached to the vibration source or to a bony spot on a person’s body. This device measures displacement acceleration in one or more dimensions. The most common descriptive measure of vibration is the root mean square (RMS) value:

$$RMS=\sqrt{\frac{1}{T}{\displaystyle \underset{0}{\overset{T}{\int }}{x}^2\left(t\right)}dt},$$

where x(t) is displacement along a particular dimension (usually specified as X, Y, or Z in three-dimensional space) as a function of time. RMS is, roughly, the square root of the average squared displacement for a fixed interval of time T.

An accelerometer should be as small as possible and should be sensitive to the ranges of acceleration and frequencies expected from the vibration source.

Operators who work with powered equipment, such as heavy vehicles or hydraulic devices, experience body vibrations for extended periods of time. As with any repetitive motion, this extended exposure can be detrimental to the health of the operator. The presence of vibration also has the more immediate effect of degrading an operator’s performance by interfering with her motor control. When we are evaluating vibration, we make a distinction between whole-body vibration and segmental, or hand-transmitted, vibration applied to particular body parts (Griffin, 2006; Wasserman, 2006). We will talk about each kind of vibration separately.

WHOLE-BODY VIBRATION

Whole-body vibration is transmitted to an operator through supports such as floors, seats, and backrests. Vibration discomfort will increase as the amplitude of the vibration increases. Most people rate RMS magnitudes of 1 m/s² or larger as uncomfortable (Griffin, 2006). However, RMS is not the only important factor in determining whether a vibration is uncomfortable. Discomfort will increase with increasing exposure times, so even minor vibration may become intolerable if a person has to endure it long enough. The frequency of the vibration is also important. Every object, even the human body, has a resonant frequency. If vibration is transmitted to an object at a frequency near the object’s resonant frequency, then the object will vibrate with amplitude higher than that of the vibration source. For the human body, the resonant frequency is approximately 5 Hz. This means that frequencies in the neighborhood of 5 Hz can have an even more damaging effect on a person’s body than frequencies outside of this neighborhood.

As we mentioned already, vibration occurs in all three dimensions. How can we assess vibration and predict whether or not it will cause discomfort? One study looked at assessment of vibration both in the laboratory and in the field (Mistrot et al., 1990). In the lab, they exposed people to vibration in either one or two axes of motion and asked them to rate their discomfort. In the field, several professional truck drivers drove a truck with different loads, at different speeds, over good or poor sections of road. They measured vibration in all three directions by putting accelerometers in the driver’s chair, and the drivers estimated their degree of discomfort. The researchers compared the drivers’ judgments of discomfort with those from the people in the lab, and concluded that discomfort is best predicted using the RMS of the displacement in each axis. That is, there is no single direction of vibration that will produce discomfort.

Whole-body vibration interferes with vision and manual control. It also can have health effects, particularly lower back pain and damage to the lumbar region of the spine, for individuals who are exposed to whole-body vibration on a regular basis. This would include truck drivers, helicopter pilots (Smith, 2006), and operators of heavy construction equipment (Kittusamy & Buchholz, 2004). Helicopter pilots experience a lot of vibration through their seats (Bongers et al., 1990). Relative to other nonflying U.S. Air Force officers, helicopter pilots are more prone to both transient and chronic back pain. The pilots’ degree of pain was related to the amount of vibration they experienced and their age. The older the pilot, and the more vibration to which they were exposed, the higher was the prevalence of chronic back pain.

One way to reduce whole-body vibration for vehicle operators is by redesigning the operator’s seat. Designs that minimize the contact between the operator and the seat will reduce whole-body vibration. One study showed that a new car seat that tilted the back of the seat down to minimize seat contact, and included a padded protruding cushion for increased lumbar support, decreased the amplitude of whole-body vibration by about 30% (Makhsous, Hendrix, Crowther, Nam, & Lin, 2005). An alternative to reducing contact with the seat is to construct seats with suspension systems that counteract unwanted vibration from the road, much as in noise-reduction headphones. These kinds of seats reduce driver fatigue and should reduce the risk of musculoskeletal disorders for drivers of trucks, vans, buses, and tractors (Wang, Davies, Du, & Johnson, 2016).

SEGMENTAL VIBRATION

Segmental vibration to the hand and arm occurs while using power tools. We will talk mostly about hand-arm vibration, but in some situations a human factors expert may need to deal with head-shoulder and head-eye vibrations. Recall that the resonant frequency of the human body is around 5 Hz. However, the human arm has very little resonance. This means that most vibrations are absorbed into the hand and transmitted up the arm. As vibration frequency increases, less of the vibration is transmitted up the arm (Reynolds & Angevine, 1977). For vibrations of approximately 100 Hz, the entire vibration is absorbed by the hand.

Segmental vibration can cause a person to misperceive the movements and location of the vibrated segment, which in turn can lead to inaccurate aimed movements and possibly accidents (Goodwin, McCloskey, & Matthews, 1972). Vibration applied directly to the tendon of one of the major elbow muscles, either the biceps or the triceps, will induce a reflexive movement of the arm. If you ask a person to match the movement of the vibrated arm by moving his other arm, the mismatch between what the person feels and what is actually happening will be apparent. The person will move his arm too much and in the wrong direction—the direction that his arm would have moved if the vibrated muscle were being stretched. Vibration feels like muscle stretch, which results in misperception of the movement at the elbow. We can do this for other joints as well, which means that vibration could severely hamper motor performance, particularly when a person can’t see or doesn’t look at the vibrated segment to verify how it is moving.

Long-term use of vibrating power tools can result in a cumulative trauma injury called vibration-induced white finger, Raynaud’s disease, or hand-arm vibration syndrome (HAVS; Griffin & Bovenzi, 2002; Poredos, 2016). HAVS arises from structural damage to blood vessels and nerves and is characterized by extreme numbness and intermittent tingling in the affected hand or finger. In later stages of the disease, the finger alternates between blanching (or whiteness) and cyanosis (or blueness), which is symptomatic of interruptions in the blood supply to the finger. The interruption may ultimately lead to gangrene, requiring amputation of the affected finger (or fingers).

The amount of exposure that a person can tolerate is a decreasing function of the intensity of the vibration. Figure 17.14 shows how many people will develop blanching symptoms after exposure to different magnitudes of vibrations (RMS on the abscissa) for extended periods of time (years on the ordinate). Most hand-arm vibration data come from men. Women show greater sensitivity to and discomfort from vibration, which means that we need to be careful to consider possible gender differences when assessing segmental vibration (Neely & Burström, 2006).

FIGURE 17.14Years of exposure before developing symptoms of advanced stages of vibration-induced white finger, as a function of the intensity of vibration.

HAVS can be aggravated by many factors. For example, a person who uses a tight grip and works in the cold, which causes the arteries to constrict, may develop HAVS more quickly. Also, some vibration frequencies are more problematic than others. In particular, exposure to vibrations between 40 and 125 Hz increases the likelihood that a person will develop HAVS (Kroemer, 1989).

THERMAL COMFORT AND AIR QUALITY

The climate of a working environment usually refers to the temperature and relative humidity of people’s surroundings. There are some workplaces where it is easy to maintain a normal temperature and humidity. However, there are other workplaces where this is not possible. A frozen-food warehouse cannot be kept comfortably warm; a tent in the desert cannot be kept comfortably cool. Extremes of temperature and humidity can severely restrict a person’s capabilities, diminishing his or her stamina, motor function, and overall performance.

To evaluate the climate in a workspace, we often refer to a comfort zone (Fanger, 1977). A comfort zone is a range of temperature and humidity that people will find acceptable given restrictions imposed by the tasks they are trying to perform, their clothing, air movement, and so forth. Figure 17.15 plots the comfort zone for moderate air speeds (0.2 m/s), light work, and light clothing. The zone is shown as a dashed rectangle in the center of the temperature–humidity range. For this zone, dry-bulb temperature (that measured by a typical thermometer) varies from 19°C to 26°C (66°F–79°F), and relative humidity varies from 20% to 85% at the extremes (Eastman Kodak Company, 1983). Under the specific conditions being depicted, people should be comfortable as long as the combination of temperature and humidity remains inside of the comfort zone.

FIGURE 17.15The comfort zones for winter and summer.

A person’s impression of comfort will be influenced by several factors. Heavy work will shift the comfort zone to lower temperatures. Workers do not often perform heavy work continuously throughout a shift, and not all workers on a shift will be performing heavy work. This means that the temperature in the work area must be a compromise between the comfort zone for sedentary work and that for heavy work. We can solve this problem by providing those workers who perform less strenuous tasks with warmer clothing, such as sweaters or jackets. High air velocities will reduce the insulating ability of clothing and will require that we set the temperature higher. In the range of comfortable temperatures, relative humidity has only a minor influence on thermal sensation.

Fanger’s (1977) comfort zone concept is the basis for attempts to develop active customized ­thermal comfort controls (Andreoni, Piccini, & Maggi, 2006). Customized control uses sets of sensors and transducers to measure the temperature and relative humidity in a room and on the body of a person who is working in the room. The climate control system takes these measurements and computes an estimate of thermal comfort based on the comfort zone, and responds quickly to changes in climate conditions and activity levels to maintain the person’s comfort.

We can also determine discomfort zones. Discomfort will arise when a person’s body’s thermal regulatory system is strained beyond its normal bounds. This will happen for some combinations of temperature, humidity, and workload. For example, when you get very hot, you may sweat a lot. Sweating is often uncomfortable. Furthermore, your tools and controls might get slippery, and when your clothes get wet it may be harder to move. Your mental acuity may suffer (Parsons, 2000), and your dexterity as well (Ramsey, 1995). People are less able to perform tracking and vigilance tasks in temperatures of 30°C–33°C (86°F–91°F) or higher.

Hot environments require managers to implement certain work practices to prevent heatstroke or hyperthermia. Workers must be provided with ample water and a cool area in which to rest. They must also be trained to recognize the symptoms of hyperthermia, and they must be given enough time to adjust to the heat when they first arrive on the job, or when they return after vacation or a leave of absence.

A person’s performance will deteriorate in the cold, too. Manual dexterity will deteriorate as a consequence of physiological reactions such as stiffening joints (Marrao, Tikuisis, Keefe, Gil, & Giesbrecht, 2005; Parsons, 2000). Also, cold environments require additional clothing, which will restrict a person’s range and speed of movement. We may need to restructure some tasks to accommodate the workers’ decreased mobility. We also need to be aware that the work environment may be hazardous for someone wearing bulky clothing: some open machinery may allow fabric to become entangled. Drafts will increase the workers’ discomfort, and so we need to make every effort to eliminate them and to provide a source of radiant heat. An increased workload will also make the cold environment more tolerable. As with excessively hot environments, workers must be trained to recognize the symptoms of hypothermia and frostbite.

Both extreme cold and extreme heat can have deleterious effects on a person’s ability to perform complex tasks (Daanen, Vliert, & Huang, 2003; Pilcher, Nadler, & Busch, 2002). Cognitive task performance can be impaired up to 14% for temperatures less than 10°C (50°F) or greater than 32°C (90°F). Similarly, driving performance decreased by 13%–16% in hot and cold temperatures. These findings imply that environmental and human factors engineers need to make the greatest effort possible to maintain moderate temperatures in the workplace, because this has direct implications for safety.

Apart from temperature and humidity, we often need to be concerned about indoor air quality. Usually, we will focus our attention on the presence of gaseous and particulate pollutants in the air, concentrations of which can build up to be many times higher indoors than outdoors.

Pollutants can be classified into three categories (ASHRAE, 1985):

1.Solid particulates, such as dust, pollen, mold, fumes, and smoke

2.Liquid particulates in the form of mists or fogs

3.Nonparticulate gases

To evaluate the air quality of an environment, we measure each of these categories of pollutants. If we find high levels of any pollutant, we have to determine its source. Some common sources of indoor pollutants include living organisms (pets, rodent and insect pests, bacteria, and mold), tobacco smoke, building materials and furnishings, central heating and cooling systems, chemicals used for cleaning, copy machines, and pesticides (U.S. Environmental Protection Agency, 2017).

Pollutants are spread from their sources by way of air movement, which is wind in the outdoor environment and the ventilation within an indoor environment. Because ventilation systems bring in air from the outside, the source of an indoor pollutant can be from either inside or outside a building. Poor ventilation can create conditions in which molds and fungi flourish (Peterman, Jalongo, & Lin, 2002). Molds can cause allergies and (depending on the type of mold) can be toxic. Molds often grow in damp areas, such as ceilings. Musty smells may give away the presence of mold. Air conditioning cooling towers can also harbor mold and bacteria, such as the bacteria responsible for Legionnaire’s Disease, and spread those bacteria through the ventilation system.

Poor air quality can also have a negative effect on performance. In one study, the level of air pollution was manipulated by introducing or removing a pollution source, an old carpet (Wargocki, Wyon, Bake, Clausen, & Fanger, 1999). The carpet had previously been removed from an office building for having an unpleasant odor and irritating employees’ eyes and throats. Participants were exposed to the pollution source or its absence for 265 minutes, unaware of the condition, since the carpet was placed behind a partition. During this period, the participants performed tasks simulating office work and filled out assessments of perceived air quality. Headaches were reported as greater when the pollutant was present, and the air quality was rated as worse than without it. The participants typed 6.5% more slowly when the pollutant was present than when it was not, consistently with reported lower levels of effort. Thus, the discomfort caused by poor air quality can have a negative impact on performance and productivity.

When many occupants of a building experience recurring respiratory symptoms, headaches, and eye irritation, they are said to suffer from nonspecific building-related symptoms, or sick building syndrome (Norbäck, 2009; Runeson & Norbäck, 2005). The syndrome is controversial, because it is a function of several medical, psychological, and social factors (Thörn, 2006), but it can have a significant negative impact on the people in the affected building (Söderholm, Öhman, Stenberg, & Nordin, 2016). Sick building syndrome is blamed on the tightly sealed buildings that were constructed beginning in the late 1970s to conserve energy. These buildings had minimal ventilation from the outside, resulting in buildups of pollutants within the building. Perhaps as many as 30% of the buildings worldwide that were built or remodeled during this period could be diagnosed with sick building syndrome (World Health Organization, 1984). Sick building syndrome is corrected by improving indoor air quality.

We can improve air quality by one of two methods. We can use devices like high-energy particulate absorbing (HEPA) filters in air purifiers. These filters remove over 99% of particles of 0.3 micrometers (microns) diameter, and larger and smaller particles are filtered even better. (For reference, the HIV virus is 0.1 micron in diameter). Alternatively, the contaminated air can be diluted with outdoor air by increasing ventilation rates (Cunningham, 1990), assuming that the outdoor air is not also polluted. All ventilation systems work by bringing outdoor air inside the building, but different buildings will require higher or lower rates of air circulation. In the U.S., state building codes will state the amount of outdoor air required for specific applications, such as the combustion of wood, dry cleaning, painting, hospitals, and so forth.

STRESS

Stress is a physiological and psychological response to unpleasant or unusual conditions, called stressors (Sonnentag & Frese, 2003). These conditions may be imposed by the physical environment, the task performed, one’s personality and social interactions, and other stressful situations at home and at work. Although specific stressors, such as temperature extremes, produce specific physiological responses in a person, they all cause the same nonspecific demand on the body to adapt itself. This demand for adaptation is stress. Acute stress associated with immediate events can be intense and affect performance; chronic stress over a period of time can have harmful physical, as well as psychological, effects on a person.

GENERAL ADAPTATION SYNDROME AND STRESSORS

In 1936, Hans Selye first characterized stress as a physiological response. He noticed that rats injected with different toxic drugs exhibited many of the same symptoms even though the drugs were different. He also discovered a characteristic pattern of tissue changes in the adrenal and thymus glands, and in the lining of the stomach wall taken from sick rats. Sick rats, or stressed rats, had swollen adrenal glands, atrophied thymus glands, and stomach ulcers. These three symptoms are the General Adaptation Syndrome (Selye, 1973). This syndrome is characterized by stages of physiological responses of increasing intensity.

The first stage in the syndrome is the alarm reaction, which is the body’s initial response to a change in its state. It is characterized by discharge of adrenaline into the bloodstream. If the stressor inducing the alarm reaction is not so strong that the animal dies, then the body enters the second, “resistance,” stage. In this stage, adrenaline is no longer secreted, and the body acts to adapt to the presence of the stressor. As exposure to the stressor continues, the body enters the final “exhaustion” stage, in which its resources are depleted and tissue begins to break down.

Stress is a function not just of physiological factors, but of psychological factors as well. Probably the most important factor is how a person appraises or construes his or her situation (Lazarus & Folkman, 1984). The person appraises the harm that has already occurred, the threat of harm that may take place in the future, and the available resources for dealing with the stressor. On the basis of this appraisal, the person will decide whether the stressful environment is merely unpleasant or intolerable, and then how he or she will react to the stressor. The appraisal can be affected by such things as the degree of control that the individual has over the situation and his or her understanding of why situations are as they stand.

Extremely high stress can severely impair a person’s ability to make decisions, particularly if the person feels that he is under time pressure. In such situations, he may react in a way that is called hypervigilance (Janis & Mann, 1977). Hypervigilance is a panic state in which his memory span is reduced and his thinking becomes overly simplistic. He may search frantically and haphazardly for a solution to a problem and fail to consider all of the possible solutions. In an attempt to beat a decision deadline, he may make a hasty, impulsive decision that has some promise for immediate relief but also has negative longer-term consequences.

Hypervigilance may contribute to some of the errors that people make in emergency situations. Emergencies are characterized by acute stress, which is induced by a sudden potentially life-threatening situation for which a solution must be found quickly. Hypervigilance has been linked to incidents of unintended acceleration: The panic induced when a person’s car is rocketing out of control reduces her ability to detect that her foot is on the accelerator instead of the brake pedal.

There are three classes of stressors: physical stressors, social stressors, and drugs (Hockey, 1986). We can also distinguish between external versus internal sources of stress and transient versus sustained stress. External stressors arise from changes in the environment, such as heat, lighting level, or noise, whereas internal stressors arise from the natural dynamics of a person’s body. Transient stressors are temporary, whereas sustained stressors are of longer duration.

Figure 17.16 shows the relationship between different stressors and a person’s internal cognitive states. The person is designated by the larger broken box, and his internal cognitive states by the smaller broken box within it. Drugs and physical and social factors provide external stress, whereas cyclical changes (such as a woman’s menstrual cycle) and fatigue provide internal stress. Physical stress is caused by annoying and uncomfortable environmental conditions of the type discussed in this chapter. Physical stressors directly influence the stress state of the individual, although their effects are mediated to some extent by the person’s cognitive appraisal of the situation. Physical stressors also can produce fatigue.

FIGURE 17.16The relation between stressors and internal states.

The influence of social stressors, such as anxiety about evaluations of one’s performance and incentives, is mediated by cognitive appraisal. In response to stress, a person may try to regulate the state of her body by taking antianxiety drugs. Thus, drugs have their primary effect on her stress state. They also can influence the person’s level of fatigue, which in turn affects her stress state.

Fatigue is the wide range of situations in which a person feels tired. It can be caused by excessive physical and mental workloads and loss or disruption of sleep. Fatigue results in feelings of not only tiredness but also boredom. Cyclical stressors are those involving natural, physiological rhythms. These stressors are usually studied by investigating performance when rhythms are disrupted, for example, by shift work or jet flight. High fatigue and disruptions of circadian rhythms (see Chapter 18) increase stress.

Table 17.3 gives a summary of the different classes of stressors and the locus of their effects. There are large differences in the extent that different people are susceptible to stress. The same stressor applied to two different people may have different effects. Moreover, the effect of a given stressor may vary depending on what a person is trying to do. The level of stress induced by a particular variable (e.g., cold temperature) may not be as great with an undemanding task as with a demanding task. Note also that the effect of a particular stressor on the stress state may be larger when other stressors are present.

OCCUPATIONAL STRESS

The term occupational stress specifically refers to stress associated with a person’s job (Gwóźdz, 2006). Healthcare workers are particularly susceptible to this kind of stress. For example, Marine, Ruotsalainen, Serra, and Verbeek (2006) noted:

Healthcare workers suffer from work-related or occupational stress often resulting from high expectations coupled with insufficient time, skills and/or social support at work. This can lead to severe distress, burnout or physical illness, and finally to a decrease in quality of life and service provision. The costs of stress and burnout are high due to increased absenteeism and turnover. (p. 2)

This quote demonstrates how stress in the work environment can arise from the physical and social environment, organizational factors, and a person’s tasks, but, in addition, a person’s personality attributes and skills are important factors (Smith, 1987).

Environmental sources are, as we have discussed, the climate, lighting, and so on in the workplace. Physical environmental stressors are more of a factor for manual laborers than for office workers and managers. Organizational factors involve job involvement and organizational support. For example, an autocratic supervisory style can lead to a person’s job dissatisfaction and, hence, increased stress. Lack of performance feedback or continually negative feedback can also be stressful. Workers in an organization that allows employees to participate in decisions that impact on their jobs will experience less stress than workers in an organization that does not. Opportunities for career development also serve to lessen occupational stress.

Job-task factors influencing stress include high mental and physical workload, shift work, deadlines, and conflicting job demands. To some extent, an individual must be “matched” to certain jobs (Edwards, Caplan, & van Harrison, 1998). Training must be appropriate, the job must be acceptable to the individual, and the individual must have the physical and mental capabilities necessary to perform the job. The degree to which a worker is not well matched to his job in training, desire, and capability in part determines the level of stress that he will experience.

There are several types of intervention that can help relieve occupational stress (Kivimäki & Lindstrőm, 2006). Those that focus on the individual include stress management training, in which the person learns stress reduction and coping strategies such as muscle relaxation. Cognitive-behavioral interventions have the goal of changing a person’s appraisal of the situation. In the case of a single traumatic event, like the accidental death of a co-worker, debriefing programs conducted within a day or two after the event can minimize stress. For the workforce as a whole, organizations may also provide employee assistance programs, promote healthy work organizations, and institute ergonomic improvements. Job redesign and organizational change, topics of the next chapter, can also be effective tools for reducing occupational stress.

Some work environments, such as a space station (see Box 17.1) or an Antarctic research station, are “contained”: a person can’t leave the work environment because it is the only environment that supports life. Forced containment restricts the actions that a person can take to reduce stress, and this restriction introduces stressors of its own. These include (1) the surrounding hostile environment, (2) a limited supply of life-supporting resources, (3) cramped living spaces and enforced intimacy, (4) the absence of friends and family, (5) few recreational activities, (6) an artificial atmosphere, and (7) an inability to leave the contained environment (Blair, 1991).

The stress of a contained environment manifests itself in several ways. A person may experience increased appetite and weight gain, as eating becomes very important as entertainment. Her decreased activity and the loss of light and dark cycles disrupt her sleep patterns. Because of her enforced proximity with other people in the environment, her sleep/wake cycle can be very disruptive to others with different sleep/wake cycles. Anxiety and depression are common, and her sense of time may be distorted.

Because these stressors cannot be removed from a contained environment, the best way to control stress in such environments is through careful screening of the applicants. People should be selected who adapt well and are not unduly affected by the stressors induced by the contained environment. Blair (1991) describes a good candidate as one whose predominant interest is in work and who is comfortable with, but has no great need for, socializing. He describes the best candidates for work in these environments as “often not very interesting people.”

SUMMARY

Human performance, health, and safety are not determined solely by the design of displays, controls, and the immediate work station. Additionally, the larger environment in which a person lives and works makes a difference between tolerable and intolerable working conditions. The goal of environmental ergonomics is to ensure that engineers appropriately consider the physical environment when designing workspaces. Some critical factors include appropriate illumination, noise levels within tolerable ranges, protection against extreme noise and vibration, task-appropriate climate, and high air quality.

Inadequate environmental conditions are major contributors to stress. Stress is also produced by a variety of other factors, including the social environment, task demands, and long-term confinement. High levels of stress can result in illness and poor performance. By selecting candidates using appropriate screening methods and designing the environment and tasks to minimize stress, we can keep stress within acceptable limits.

RECOMMENDED READINGS

Behar, A., Chasin, M., & Cheesman, M. (2000). Noise Control: A Primer. San Diego, CA: Singular.

Boyce, P. R. (2014). Human Factors in Lighting (3rd ed.). Boca Raton, FL: CRC.

Harrison, A. A. (2001). Spacefaring: The Human Dimension. Berkeley, CA: University of California Press.

Kroemer, K. H. D., & Kroemer, A. D. (2001). Office Ergonomics. London: Taylor & Francis.

Mansfield, N. J. (2005). Human Response to Vibration. Boca Raton, FL: CRC.

Oborne, D. J., & Gruneberg, M. M. (Eds.) (1983). The Physical Environment at Work. New York: Wiley.