Products are made to be used and as such are expected to function, to be usable and to be safe. It is only possible for designers to achieve this in their designs if they have some understanding of both the physical and the psychological human characteristics. The tools and data of ergonomics and their use, in particular those of anthropometry (the measurement and collection of data of body dimensions), are an essential part of industrial design education. Students need to understand both how to apply the data appropriately and the limitations of the use of such data.
The anthropometric databases which are used have traditionally been available in tabular, pictorial/diagrammatic form (eg Pheasant 1988) and also as design recommendations (eg.Dreyfuss, H.,1967; Diffrient, Tilley & Bardagjy, 1978; Diffrient, Tilley & Harman, 1981a & b). They are all designed to provide information which allows the designer to base their designs on apparent fact; the size of particular body dimensions of very small (2.5 or 5th percentile) & very large (95th or 97.5 percentile) females & males.
There are both content and presentation differences in the different forms: the tabular and pictorial forms both provide data which the deigner is responsible for selecting appropriately for his requirements (ie. in terms of age, gender, nationality, clothing etc. as well as choosing the correct body dimensions for the particular design); the designer is given the responsibility of ensuring the data chosen is the correct data. Recommendations, however, appear in a form where those assumptions have already been made on behalf of the designer and consequently must be regarded as more general, or in some cases more specific, guidelines from which to specify a design. All provide a useful starting point on which to base the dimensions of any design but are limited in their usefulness in that they they provide no further information about the way in which any individual may interact with or feel about a product.
These types of information are now available relatively cheaply in similar forms as computer databases (eg. Ergobase - available for a number of years and People Size - a new package only recently available) and also as ergonomics evaluation tools (eg. Mannequin) which remove the necessity for extensive mock-up stages in the design process and are ideal for exploratory investigations in the early stages of design. All are available to Transport and Product Design students in Coventry School of Art & Design.
It has, however, been observed that students are reluctant to use these facilities; in some cases at all and in others in an appropriate fashion. Ergobase has tended not to be used, Peoplesize as yet remains an unknown quantity and Mannequin is used mainly as a visualisation tool only. The facilities are made known to students in both Ergonomics and Computing modules and they use extensively the traditional forms of data/recommendations. It is thought that there are a number of reasons for this:-
The discussion of these issues is based upon a combination of interviews with MA students, following a practical session where they were required to use all three packages to produce recommendations for a driving package for a world population and staff opinion. Ergobase and Peoplesize are compared directly whilst Mannequin, because of the different nature and potential use of the package, is considered seperately.
Ergobase is easy to use for the novice user but the data appear in a dry scientific tabulated form described in anthropometric terminology; explanatory diagrams are available but they are difficult to comprehend and do not appear on the screen automatically. This causes a problem if the user is unsure of anthropometric terms (designers often are) and is designing from scratch. Resorting to a traditional data source, where the data is explained both pictorially and verbally, is often easier and more satisfying.
In contrast Peoplesize offers a pictorial/diagrammatic representation of all measurements that are contained within the database as a starting point; the user can choose from the representation a measurement which looks appropriate, the measurement is then explained in order that the user is able to be sure that it is the correct one before the data is given.The students found this generally much easier to interact with, as relatively naive users, than Ergobase.
In the case of Manneqin the user is able to choose a 3D "person" of any percentile value and alter body dimensions to a different percentile value as required. The constructed "person" can then be used to assess the environment for that size of user. The "person" is easy to construct and changing a particular dimension is also straight forward. The package allows the assessment of worst case situations eg. a driving package for the individual with very long legs (seat in rearmost position) and relatively short arms (unable to reach the steering wheel adequately?). The students enjoyed the ability to produce "people" of different body sizes but found designing the driving package itself very difficult because the anthropometric data was opaque. Building the environment was easier using a traditional anthropometric database as a data source or basing the design on recommendations eg Diffrient et al(1978) as a starting point. Visually the package was very popular with students perceiving it as an easy and acceptable (computerised and therefore "state of the art") way of producing the human form.
It is vitally important that if one is to use anthropometric data correctly that the data source is understood. The data must be estimated from measurements of a population of appropriate age, gender, nationality, specific nature of the measurement (the specific posture adopted whilst the data was being collected - slumped or upright?) and clothing type for the product being designed. This is true whatever the form of information.
In all three packages it is not easy to identify the details of the population. In the case of Ergobase there is a wide range of ethnic origin data available but no details are given of the nature of the measurements. If one resorts to the literature accompanying the package there is a bibliography of sources but no details of how the individual sources relate to particular dimonsion data.
Peoplesize again offers a wide range of ethnic origins, gives the clothing allowances which have already been made and offers recommendations for the amount of extra space which may need to be considered for clothing. Again the literature accompanying the package offers a bibliography of sources but no details of the individual sources or the particular data sources for particular dimensions.
The same problems described above exist with the data which is used in Mannequin. These issues make appropriate use of the data at best awkward and at worst incorrect for the application.
Because the students chose to use recommendations as the source for building their driver's package in Mannequin it made them very aware that they are the easiest source on which to base a design. However designers tend to use recommendations with the assumption that they are correct although they actually require the same assessment to be made of their appropriateness as anthropometric data. For example, the Human Scale recommendations are based on US only anthropometric data and to design a vehicle for a world population also requires anthropometric knowledge of the more diminutive Japanese population; it is imperative that the designer questions the details of the recommendations before use. There are also a number of different sets of recommendations available and the information between sets rarely concurs absolutely; a number of sources should be cross-referenced before use in design. The students, although having being warned of these issues, allowed the convenience of use factor to dominate their choice of tool.
Hasdogen (1992) found both designers and ergonomists very critical of anthropometric data. The ergonomists however, are still willing to use the data as a starting point for design (accepting its limitations) whilst designers tend (because of these limitations) to not trust the data and to base their designs on internalised knowledge/ experience which they assume to be correct. One only has to look around the market place to know that products are not always designed with the user populations' needs, abilities, skills and preferences in rnind; designers do sometimes get it wrong. This informal finding concurs with the beliefs of Pheasant(1988) who lists the five fundamental fallacies, commonly held by individuals in the design profession, as:
Such beliefs clearly interfere with the process of succesful design for "human use" and also suggest that ergonomists simply hav'n't got it right where the presentation of the the tools they believe designers should use are concerned.
Whilst the students seem willing to comply with an educational requirement to use ergonomics data as an initial starting point for their designs they do not use Mannequin as an ergonomics evaluation tool. It is used as a visualisation tool ie to show a vehicle interior with passengers (any obvious ergonomics faults would be picked up eg a roofline which is too low) but not to actually evaluate a product. Mannequin is fairly powerful in that it allows a "person's view", so the visibility of controls and displays could be assessed for different percentile individuals, as well as the more fundamental issues of fit to be evaluated.
Another fundamental problem with the use of the package would seem to be that whilst students are required to use ergonomics a majority of their marks in design projects are given for other aspects of design. The package does requires some knowledge of 3 D modelling and the students find it difficult to find sufficient time within the academic timetable to acquire the skills; time is a big constraint for all but the most enthusiastic. They feel that they cannot afford the time to produce both a good aesthetic design which is well presented and do a thorough ergonomics evaluation. If they do want to conduct an exploratory (quick & dirty) ergonomics evaluation they prefer to build a basic "experimental rig" because they can also use it to investigate other subjective issues such as ease of ingress/ egress which it is difficult using a CAD package such as Mannequin. The use of such experimental rigs does indicate both a responsibility for the input of ergonomics and a sufficient level of ergonomics sElls to be able to use Mannequin if motivated. Essentially although the students feel a responsibility to get the ergonomics "right" they perceive other aspects of the design process as being much more important.
There appear to be a number of reasons why students do not take advantage of the ergonomics computer packages on offer in Coventry School of Art & Design:
The above findings pose an interesting challenge to ergonomists designing tools (computerised or otherwise) for designers. Can they get right for the designers as well as the ergonomists?
Diffrient, Tilley & Bardagjy, 1978, Human Scale 1/2/3. (Cambridge, Mass.: MIT Press).
Diffrient, Tilley & Harman, 1981 Human Scale 4/5/6 (Cambridge, Mass.: MIT Press)
Diffrient, Tilley & Harman 1981 Human Scale 7/8/9. (Cambridge, Mass.: MIT Press).
Dreyfuss, H., 1967, The measure of man: human factors in design. (Whitney Library of Design).
Hasdogen, G.,1992, The Nature and Limitations of User Models in the Household Product Design Process. Unpublished PhD thesis, The London Institute.
Pheasant, S.(1988) Bodyspace Taylor & Francis, London