ABSTRACT
Displays A display then is a device, mechanism or channel by which information relevant to the task is transmitted to the operator. It should be noted, however, that despite its name a display is not necessarily a visual device, although it commonly is. It could, for example, be auditory, such as a buzzer or bell, tactile, such as a change in textured surface, or kinesthetic, such as the vibrating stick shaker that signals impending stall in an aircraft. Functional criteria There are certain functional criteria that must be met in a display. For example, depending on the needs of the system there may be a requirement for different priorities in speed of response, or in other words, how quickly should the display reflect any change in activity, how accurate should be the presentation, or for example, how closely should it reflect the change, and how small a change should it respond to. Consequently, certain design criteria need to be established before devising or selecting the appropriate display. These include:
N ature of the variableThe particular variable that is co be displayed should be clearly established before any design is considered. For example it is no use to display a value when what is really wanted is a rate of change of that value. Total range of variablesThe range of the value to be displayed will determine the design of the display. A range involving only a small quantity will require very different treatment from one involving a large quantity. M aximum required accuracyThe maximum required accuracy determines the precision of the scale. M aximum speeed of transferThe maximum speed of transfer determines how rapidly the required information must be presented to the display. M aximum equipment errorAll equipment has a certain intrinsic error. The less the acceptable error the more complex and more expensive the display system may have to be. D istance between display and userThe distance between display and user largely determines the size and clarity of the visual components. On this basis one can begin to determine what kind of display is required to meet a given purpose. Types o f display Displays can be real or artificial, static or dynamic. Real displaysAlthough it is not commonly thought of as such, in a real display the environment itself provides the information directly. For example, by this definition, the windshield of a car is indeed a display and meets the criteria for a real display. Artificial displaysArtificial displays provide a symbolic representational value of the information of interest, as for example the speedometer of a car. In the common design of a car speedometer the needle of the display points to a number which the driver interprets in terms of speed. Static displaysStatic displays are not instruments and do not change with conditions. They give information by their existence and by the interpretation on
the part of an operator of the information contained therein. The adjustment setting for a lathe, for example, which is often attached as a plate or etched into the metal, is a static display. D ynamic displaysA dynamic display is one which responds to the actual represented value, and changes as that value changes. An electrical voltmeter, for example, like a speedometer, is both an artificial display and a dynamic display. Categories o f display Q ualitative displays
These are used to distinguish between a small number of discrete conditions such as on and off, open and shut, normal and abnormal, and so on. There are three distinguishable types, which have in fact already been noted in passing. These are: Auditory displaysAuditory displays include bells, buzzers, beeps, and the like. Their main advantage lies in their pervasiveness. One is warned without necessarily paying attention. They have the disadvantage, however, of being masked by competing noise. In this connection, it should be recognized that warbling sounds in short bursts are more attention-getting than continuous sounds of the same pitch. V isual displaysVisual displays include lights, colours, and even the position, for example, of a lever which has been moved. They can be made distinctive in a variety of ways including varying sizes and shapes according to demand. Colour can be useful provided that one remembers that some 8 per cent of the population may be to a greater or less extent colour blind. Varying intensity of lighting can be used to emphasize different characteristics, but this approach is only of real usefulness when the light is changing. The great disadvantage of visual displays of course is the fact that they have to be looked at. It might be remembered, however, that flashing lights tend to attract the peripheral vision and direct the attention of the operator towards the source of the flashing. T actile displaysTactile displays in this connection include kinesthetic. While texture has been used or at least incorporated into other forms of tactile display they are mostly represented by some form o f‘positioning’ which is perceived by the operator’s hand or foot. For example, one can tell by ‘feel’ whether
the handbrake of a car is on or off, or whether the gear lever is in first gear or reverse, and so on. Signal and warning lightsProbably one of the most critical uses of qualitative displays lies in the provision of signal and warning lights. These are used to attract attention, to denote alarm, or simply to indicate the status of a system or component.For a signal to be detectable it must meet certain criteria. The absolute threshold of visual detectability depends on the size, that is, the angle subtended at the retina, the luminance, or the amount of light emitted, and the duration of exposure. The threshold of luminance in fact varies inversely with the exposure time (McCormick and Sanders (1982)).The addition of colour to the signal can change its detectability, but where there is a high absolute brightness of the signal, with a high brightness contrast against a dark background, then there is nothing to be gained by adding colour. Where there is a low brightness contrast the fastest response to colour, other things being equal, is found with red, green, yellow and white, in that order (Reynolds, White, and Hilgendorf, 1972).Flashing lights are more readily detectable than steady state lights, hence flashing lights should be reserved for emergency purposes.As the flash frequency is increased, eventually the observer perceives the flashing light as being continuous. The frequency at which this occurs is referred as the fl ic k e r fu s io n fr e q u e n c y , which is generally considered to be about 30 flashes per second. The recommended flash rate for emergency flashers is 3 to 10 per second with a duration of not less than 0.05 seconds per flash (Woodson and Conover, 1964).The background against which a signal is presented can modify its detectability. The worst case is found, as might be expected, where there is a flashing signal against a flashing background; and, of course, the best case occurs where the signal is flashed against a steady background. Intermediate to these is a steady signal on a steady ground.It is also preferable to have one warning light rather than multiple lights. Multiple lights can be very confusing, the more so under emergency conditions. Where several warnings are required at the same time it may be desirable to have a single warning light with an annunciator panel to indicate to what the warning refers. The intensity of a light should be twice that of the immediate background and its location should be within 30 degrees of the operator’s normal line of sight (McCormick and Sanders, 1976). P seudoquantitative displays
As the name would imply, pseudoquantitative displays provide non-numerical comparative information. They are useful in determining the
status of a function which has a limited number of predetermined ranges, for example high, medium, or low; or for selecting and maintaining some desirable range, or for observing trends in the activity of a function.Commonly a dial with a pointer is used to present the information, the dial in this case being uncalibrated and without numbers. The degree of change can be represented on the dial by position, for example by having segments of the dial marked high, medium, or low, or perhaps by colour coding with the background of the dial passing from green for safety, through yellow, to red for danger.In some cases, particularly in the monitoring of multiple machines, it may be necessary to have numerous dials. The most effective monitoring in this situation is undertaken by organizing banks of dials, for example in 3 x 4, or 4 X 4 matrices, all referring to the same type of function. Perception of change is then based on monitoring the totality of the pattern presented by the multiple pointers. If normal is defined by having a pointer at, say, 12 noon or 9 o’clock, then any deviant pointer will be readily observed (Dashevsky, 1964). The observation can be enhanced by extending graphic lines from the tip of one pointer to the base of another. In addition, sub-patterns of dials can also be defined. Figure 9.2 illustrates these concepts. Q uantitative displays
Quantitative displays present numerical information in a visual medium. There are two basic types, namely analogue which present the information
(0)
(d)
(b)
by means of a pointer and a numerical scale, and digital, which present the information in the form of changing numbers.The analogue display most commonly uses a pointer moving against a fixed scale, although it may also be used with a moving scale operating behind a fixed pointer. It is commonly easier to assimilate information from an analogue display than from a digital although it is less precise. In fact assimilation of precise information is faster from a digital than an analogue (Zeff, 1965). An analogue display also shows trends, or rate of change, more readily. Heglin (1973) lists certain characteristics which should be borne in mind in the selection of analogue displays noting that in general people prefer a moving pointer to a fixed pointer. R epresentational displays
A representational display presents a picture or a model of the system to be controlled, as for example a graphic and dynamic two-dimensional model of a rail marshalling yard, or the process control system of a refinery. The design in these cases is specific to the needs and may involve various qualitative and quantitative sub-displays integral with the total pattern. Scale and dial design criteria In selecting a scale it is necessary firstly to determine the range of scale to be covered. Within that range it is also necessary to determine the reading precision required, which in turn determines the number and positioning of the scale markers. Thirdly it is necessary to determine the format and terminology of the readout, for example, velocity, pressure, voltage, and so on, and whether the format should be in direct readout or in computed readout. For instance, it might be more useful to have a scale in percent of some predetermined value rather than in the actual value.In determining the design of the scale itself, there are three areas for special consideration. These are the number of marked divisions, the size of the estimated subdivisions, and the organization and structure of the scale. N u m b e r o f m arked divisions
The number of marked divisions turns out to be a compromise between the need for speed of reading and the need for accuracy. Where there are too many divisions the reading is slow but more precise; where there are too few reading may be more rapid but tends to be less precise. The compromise is the smallest number compatible with the requirement, allowing the operator where necessary to subdivide by eye.
