In an experimental research study, Lawson (1984) compared the ways in which designers (in this case architects) and scientists solved the same problem. The scientists tended to use a strategy of systematically trying to understand the problem, in order to look for underlying rules which would enable them to generate an optimum solution. In contrast, the designers tended to make initial explorations and then suggest a variety of possible solutions until they found one that was good, or at least satisfactory.
The evidence from the experiments suggested that scientists problem-solve by analysis, whereas designers problem-solve by synthesis; scientists use 'problem-focused strategies' and designers use 'solution-focused strategies'.
Some other studies have also suggested that designers tend to use conjectures about solution concepts as the means of developing their understanding of the problem. Darke (1984) found that designers impose a primary generator onto the problem, in order to narrow the search space and generate early solution concepts.
This primary generator is usually based on a tightly-restricted set of constraints or solution possibilities derived from the design problem. Since 'the problem' cannot be fully understood in isolation from consideration of 'the solution', it is natural that solution conjectures should be used as a means of helping to explore and understand the problem formulation. Making sketches of solution concepts is one way that helps the designer to identify their consequences, and to keep the problem exploration going, in what Sch6n (1983) called the 'reflective conversation with the situation' that is characteristic of design thinking.
Drawing and sketching have been used in design for a long time, certainly since long before the Renaissance, but the period since that time has seen a massive growth in the use of drawings, as designed objects have become more complex and more novel. Many of Leonardo da Vinci's drawings of machines and inventions
from the Renaissance period show one of the key aspects of design drawings, in terms of their purpose of communicating to someone else how a new product should be built, and also how it should work. Some of Leonardo's design drawings also show how a drawing can be not only a communication aid, but also a thinking and reasoning aid. For example, Leonardo's sketches for the design of fortifications (Figure 7) show how he used sight-lines and missile trajectories as lines to set up the design of the fortifications, and how his design thinking was assisted by drawing. In such drawings we see how the sketch can help the designer to consider many aspects at once; we see plans, elevations, details, trajectory lines, all being drawn together and thus all being thought about, reasoned about, all together.
Half a millenium later, we still see designers using essentially similar types of sketch to aid their design thinking. The early concept sketches for a house design by the contemporary architect Charles Moore (Figure 8) show similar kinds of representations as those used by Leonardo: plan, elevation and section all being
considered together with considerations of structure and calculations of dimensions and areas.
What might we learn about the nature of design thinking from looking at examples of what designers sketch? One thing that seems to appear is that sketches enable designers to handle different levels of abstraction simultaneously. Clearly this is something important in the design process. We see that designers think about the overall concept and at the same time think about detailed aspects of the implementation of that concept. Obviously not all of the detailed aspects are considered early on, because if they could do that, designers could go straight to the final set of detailed drawings. So they use the concept sketch to identify and then to reflect upon critical details, particular details that they realise might hinder or somehow significantly influence the final implementation of the complete design. This implies that, although there is a hierarchical structure of decisions, from overallconcept to details, designing is not a strictly hierarchical process; in the early stages of design, the designer moves freely between different levels of detail.
The identification of critical details is part of a more general facility that sketches provide, which is that they enable identification and recall of relevant knowledge. As the architect Richard MacCormac has said about designing, 'What you need to know about the problem only becomes apparent as you're trying to solve it.' There is a massive amount of information that may be relevant, not only to all the possible solutions for a
design problem, but simply to any possible solution. Any possible solution in itself creates the unique circumstances in which these large bodies of information interact, probably in unique ways for any one possible solution. So these large amounts of information and knowledge need to be brought into play in a selective way, being selected only when they become relevant, as the designer considers the implications of the solution concept as it develops.
Because the design problem is itself ill-defined and ill-structured,
a key feature of design sketches is that they assist problem structuring through the making of solution attempts. Sketches incorporate not only drawings of tentative solution concepts but also numbers, symbols and text, as the designer relates what he knows of the design problem to what is emerging as a solution. Sketching enables exploration of the problem space and the solution space to proceed together, assisting the designer to converge on a matching problem-solution pair. Problem and solution co-evolve in the design process.
Designers' use of sketches therefore gives us some considerable insight into the nature of design thinking and the resolution of design problems. These problems cannot be stated sufficiently explicitly such that solutions can be derived directly from them. The designer has to take the initiative in finding a problem starting point and suggesting tentative solution areas. Problem and solution are then both developed in parallel, sometimes leading to a creative redefinition of the problem, or to a solution that lies outside the boundaries of what was previously assumed to be possible.
Solution-focused strategies are therefore perhaps the best way of tackling design problems, which are by nature ill-defined. In order to cope with the uncertainty of ill-defined problems, the designer has to have the self-confidence to define, redefine and change the problem as given, in the light of solutions that emerge
in the very process of designing. People who prefer the certainty of structured well-defined problems will never appreciate thedelight of being a designer!
Engineering Design Methods
Strategies for Product Design
THIRD EDITION
Nigel Cross
The Open University, Mi/ton Keynes, UK
JOHN WILEY & SONS, LTD
Chichester- New York. Weinheim • Brisbane. Singapore. Toronto
