6.5. Future needs for research

The degree to which potential users are involved in the product design process is a crucial factor affecting the success of implementation. No implementation can be successful without the ability and willingness of the staff to co-operate. The users should be able to tailor systems to their own needs. They should have enough time and opportunity to learn and develop. The results of this study were promising in view of implementation. With the methods used here, it is easy way to involve customers in the process of evaluation and development.

However, design time and financial resources are often limited. Product development is also often strictly confidential, and involving outsiders is considered a risk for confidentiality. There are also practical limitations regarding the access to different user groups and the costs of involving large numbers of potential users in design. The need for accommodating a range of different user abilities and contexts in the design creates further challenges. Decisions also include trade-offs: a product can, for example, be made lighter with at expense of manufacturing costs. The user requirements, economic requirements, feasibility requirements and time constraints set conflicting pressures on the development team. Also, user requirements per se may be conflicting, especially in an inclusive design development work. Recognising, understanding and handling such trade-offs to maximise the success of the product is very challenging.

The methods used in this study are not the only relevant methods for the evaluation of products and systems. It would be worth while also to investigate other methods. In Japan, Kansei engineering has been found to be suitable for practical applications in such areas as industrial design evaluation in the development stage, or consumer preferences research upon the introduction of a product into the market (Nakada 1997, Jindo & Hirasago 1997). Quality function deployment (QFD) is a method that supports the process from problem identification to design specification (Roozenburg & Eekels 1995). QFD and conjoint analysis have been integrated successfully by Gustafsson (1993) and Toiviainen (1996). The method has been tailored to suit the organisational context, but may be used as a means for developing new products and modifying existing products. Berquist and Abeysekera (1996) used QFD in developing safety shoes for cold climates and found it suitable also to complement ergonomic methods. Lee et al. (1998) preferred more business-driven product development strategies, such as High Touch, as the key to product success. Fuzzy decision-making methods have also been used to solve multi-objective decision problems, when the objectives are of varying degrees of importance (Yager 1977, Hsiao 1998). Also the sociotechnical approach in system design would be worth investigating in more details as done by Clegg (2000).

At the moment, there is too much focus on ”first-time use” in the usability evaluations (de Vries et al. 1994). There is a need to move towards doing more contextual research and also to address long-term use and habitual use patterns. In the case of furniture, for example, we should spend more time on developing a more comprehensive opinion and to bring out the possible discomfort and pains that may be caused by the tasks. Contrariwise, Galer and Page (1996) concluded that, in many product evaluations, it is quite legitimate to argue that the user’s perceptions of the experience of using the product are an adequate predictor of the user’s long-term experience of the product and, therefore, of whether other, perhaps more direct, measures should be used in a complementary fashion to give a more accurate overall conclusion.

The 7-point scale used to evaluate the task-surface heights (paper II) was close to the fitting trial. Fitting trial is an experimental study in which a sample of subjects use an adjustable mock-up of a product or a workstation in order to make judgements as to whether a particular dimension is “too big”, “too small”, or “just right” (Pheasant 1996). Critical dimensions of the product will be adjusted through a range of values, and the subjects will be asked to express their preferences with respect to comfort, ease of use, and so on (Pheasant 1995). The fitting trial design would warrant more ways of evaluation of traditional ergonomics and could also be applied to usability testing. Preference and objective measures can be combined, as has been shown by the Borg-RPE (rating of perceived exertion) scale. Using this scale, exertion during dynamic work can be rated on a scale from 6 to 20, and heart rate has been fitted to be about 10 times the rating (Sanders & McCormick 1993). The idea of substituting physiological measurements by subjective ratings is attractive, as ratings do not require any instrumentation.