تويت
The goal of human factors and ergonomics is to improve the performance, reliability, and efficiency of systems in which humans operate in concert with machines, man/machine/systems (MMSs). This discipline developed rapidly during and after World War II. The life-or-death stakes and the advances in military technology made even minor improvements in military systems highly desirable and major improvements essential. In subsequent years the risks of accidents and the complexity and costs of military hardware, manned space exploration, nuclear power plants, vehicular control, and civil air transportation were behind the continued development of human factors/ergonomics as an engineering discipline. Commercial applications in product design, industrial process control, quality control, health care technology, computers, and office design have expanded this development
Human factors engineering and ergonomics draw on or interact with the theories and data of many diverse disciplines: psychology, physiology, and applied physical anthropology; aeronautical, electrical, industrial, mechanical, and systems engineering; and computer and cognitive sciences. Humans operate as sensors, information processors, actuators, and decision makers. The models used to describe these operations are determined by the MMS. For an important and extensively studied family of MMSs—the manual control of air, space, ground, or sea vehicles— the range and variability of allowable manual control is constrained, particularly when the system is operating near its stability limits. The repertory of human control dynamics behavior modes, which demonstrates the adaptive skills the operator applies to reorganizing inputs so as to maintain system performance, can then be described with the same mathematics as the inanimate system components. In applications which emphasize signal detection and decision making (as in monitoring or visual search) or estimate human reliability, performance is best described probabilistically. When the human is a significant source of energy (as in heavy industry or intense athletic activities), power output decrements over time are the main interest.
What follows presents hints at this mass of information, with some useful empirical rules, and concludes with an example of closed-loop compensatory control, the first stage in the repertory of human dynamics behavior modes.
Psychomotor Behavior
Psychomotor behavior is the activity of receiving sensory input signals and interpreting and physically responding to them. Humans can receive inputs by vision, hearing, smell, and the cutaneous senses which respond to temperature, mechanical energy, or electrical energy. Kinesthesis and the vestibular sense inform about location and position. Vision followed by hearing are the most important senses for transmitting signals carrying complex information for decisions and for control of MMSs. Signals for warning or alerting need not be complex and can be transmitted by one or a combination of the sensory channels. The choice is determined by the situation and the task being performed by the person or persons to be warned rather than by differences in modality reaction times.
The sense dominating psychomotor behavior, vision, receives light signals by either of two different retinal receptors, cones for photoptic and rods for scotopic vision, which convert optical images to signals sent to the brain. Daylight color vision and detailed visual acuity are photopic. Scotopic vision is black and white with shades of gray. Cones are concentrated in the fovea, a central area whose visual angle subtends 30 minutes of arc, and are found in very small numbers elsewhere on the retina. The distribution of rods extends to the periphery of the retina but not to the fovea. Their concentration is greatest at about 20° from the fovea.
After adaptation in the dark for 5 min, the cones are at their maximum sensitivity to low light levels below which color vision is lost. After 30 min in the dark, the rods reach their maximum sensitivity to low light levels. Adapting from a dark environment to a light one takes about a minute; time to readapt to the dark can be brief if the time has been spent under red light. Red light stimulates cones, but not rods, enabling the rods to retain their previous adaptation to a low light level.
Perception of objects and events results from sensory inputs and their interpretation by the brain. Deliberately devised size and distance illusions trick the brain by manipulating expected cues for size, distance, and perspective. Unintended illusions or misleading perceptions can arise when the brain relies on inappropriate sensations or expectations. The process of perception of the size and distance of an object begins when light from the object passes through the observer’s pupil and lens. An image is focused on the retina by the action of the ciliary muscles, which can change the accommodation of the lens. Estimates of an object’s apparent distance are influenced by this muscle action, by the visual cues leading to the object, and by the expectations of the viewer for the size of the object. These cues may be inadequate or they may be inconsistent with one another and as a consequence create a perception of distance which differs from reality. As an example, the vision of the driver of an automobile may be accommodated to the dashboard displays at a distance focus of 0.5 m. Refocusing on an unexpected distant object may take as long as 0.4 s. The object may be unfamiliar and visual cues obscured by darkness or fog. Even a briefly held misperception of object size or distance risks an error and an accident.
Complex psychomotor behavior may be constructed by incrementally aggregating a sequence of elemental discrete actions, by continuous closed-loop control, by complex open-loop activities, and by combinations of the above. Choice reaction time (RT) is an elemental discrete action whose duration is a function of the number of alternative choices and their probabilities. An example is a horizontal line of n signal light bulbs, below each of which is a key to be struck as quickly as possible when the corresponding bulb lights up. If n 5 1, RT would be about 200 ms for a practiced alert operator. As n increases, the operator’s uncertainty, H 5 log2 (n 1 1), which allows for no response, increases, and RT increases proportionally. This is the Hick-Hyman law
In information theory H is a measure of uncertainty or entropy in bits. H can be calculated whether the stimuli are equally likely or whether their probabilities of occurrence differ. The upper limit of H is slightly more than 3 bits. For the example given a is about 200 ms, b is about 150 ms, and H ranges from 0 to a little over 3 bits. An application is the calculation of RT differences among MMS designs for which values of a are likely to be similar.
The movement time (MT) for rapid discrete or repetitive actions, usually by the hands, is proportional to the difficulty of the task. Fitts’ law defines an index of difficulty ID in terms of target width W and movement amplitude A:
The form of Eq. (17.6.3) is an empirical description of data from a wide range of rapid movements as in operating a key pad or in sorting items into bins. In applications, d is about 100 ms and c, which depends on the movement geometry, is approximately 200 ms. Estimates of MT provide comparisons among operating procedures. The equations for RT and MT imply a stable level of skilled performance. The learning curves for psychomotor skills which depend on negative feedback are exponential functions of time. For other psychomotor tasks, skill learning follows a power function whose form is the same as the empirical learning curve developed by Wright to predict the unit cost of producing aircraft, and is
where y is the number of direct labor hours to produce the xth unit, a is the hours to produce the first unit, and b is the rate that labor hours decreases with cumulative output. For learning a psychomotor skill, a ranges from 0.2 to 0.6. For production a ranges from 0.1 to 0.5 with an expected value across industries of about 0.3. The empirical power function when fitted with a straight line on log-log coordinates smooths variability and simplifies quantifying production improvements over time. Each doubling of cumulative output results in a reduction of unit cost by a fixed percentage of its beginning cost. A progress ratio p of 80 percent means that after each doubling of the cumulative unit output, the unit cost is 80 percent of its value before doubling.
