What is creativity? Many would agree that creativity must result in a novel output, in the sense of not previously known. However, there is wide disagreement on two other potential dimensions of creativity: whether it must be intentional, and whether it must have influence or impact. We believe that creativity must be intentional. Products that are created unintentionally and that remain unrecognized, such as accidentally spilling paint on a canvas and then discarding the canvas in the trash, are not fundamentally creative. [1] Intention does not imply a linear or even a known process, only that effort is being deliberately applied. We also believe that creativity should not be defined by influence or impact; this is the role of the metrics applied after ideation. Experience would suggest that a focus on impact can restrict creativity. For example, brainstorm participants are encouraged not to criticize the ideas of others during the ideation phase.

The key observation we make here is that in an ideal creative process, the number of concepts under consideration should balloon (Figure 12.2). This idea was first articulated by Alex Osborn, the originator of the word “brainstorming,” who hypothesized that “quantity breeds quality.” [2] Perhaps quantity alone is not sufficient, but we will take this idea as a starting point here and refine it in Part 4.

There are two broad schools of thought on how to ideate many concepts: unstructured ­creativity and structured creativity.

The approach of unstructured creativity is far more prevalent. It includes brainstorming, blue sky ideas, free association, and related techniques. This school focuses on ideating without prejudices or biases from previous experiences. Edward de Bono [3] asserted that the mind is physically constrained by channels of thought and that the purpose of creativity is therefore to form new pathways through the concept-space. This approach is well illustrated using one of his creativity techniques: opening a book to a random page, blindly pointing to a word, and then attempting to link the word to the problem at hand.

Unstructured creativity is rooted in a notion of unconscious processing—an “Aha!” moment that arises seemingly out of nowhere, providing the creative thinker with a solution. Some would argue that creativity is by definition unconscious and must occur without warning. If the roots of the idea are known, or the process is defined, then it is not creative. Although this is a feasible definition, we believe it is somewhat counterproductive. It holds creativity to a standard at which it cannot be encouraged. We will try to encourage use of both conscious and unconscious processing.

History has a long list of “creative thinkers” (Einstein, Picasso, da Vinci, Maya Angelou, and so on), who in characterizations are imbued with divergent thinking, the ability to come up with many novel ideas. The counterpoint to this concept of unconscious, native creativity is structured creativity, which holds that creativity can be stimulated by analysis and that creative thinking is not fundamentally different from problem solving and ordinary thought.

Structured creativity asserts that problem analysis can be helpful in solution synthesis. Recall the corkscrew example in Chapter 7, where “cork removing” generated a narrower list of possibilities than the solution-neutral statements of “wine accessing.” We will build on this example of structured creativity with component recombination and completeness frameworks, described below.

A frequent theme in structured creativity is component recombination, the idea that the problem is decomposed into pieces, where more than one choice per “piece” is available, and then a solution is synthesized by selecting one choice for each piece. We have been deliberately vague with the pieces of the decomposition, because they can be form–function assignments (as in Figure 12.3, where a form is chosen for each of the three functions), or they can be specific functions for a given solution-neutral function. This same idea is encoded in morphological matrices (introduced in Chapter 7) and other decision support tools presented in Chapter 14.

Completeness frameworks, another form of structured creativity, use lists to stimulate ideation. For example, we could develop a list of forms of energy: linear kinetic energy, rotational kinetic energy, potential energy, chemical energy, and so on. When faced with a problem, we would ask ourselves what a concept employing each form of energy would look like. For example, many modern hybrid cars use batteries to store energy for use in propelling the vehicle. However, we could also use rotational energy, such as the flywheel used in Porsche’s GT3RS hybrid vehicle and in the 2009 Williams F1 car.

A more abstract version of a completeness framework is de Bono’s Six Hats. [3] This team-based method defines six roles for team members to play in problem solving, The structure of this approach is based on a theory that challenging different modes of reasoning leads to a more holistic evaluation of the problem and potentially spurs new solutions. Although the six hats (Managing, Information, Emotions, Discernment, Optimistic Response, and Creativity) are not proven orthogonal directions from a “brain and cognitive ­science” perspective, they illustrate a possible decomposition of team roles for working groups.

One of the most famous structured creativity approaches is known as TRIZ (Theory of Inventive Problem Solving) and was developed by Genrich Altshuller in the Soviet Navy, who reviewed 40,000 patent abstracts to define 40 principles of invention. Altshuller begins with seeming contradictions, such as “a faster train will require a more powerful engine, but a more powerful engine will be heavier . . . (thus reducing the acceleration gains to be had from the higher power)—we wish to go faster for the same weight.” These contradictions are resolved using the set of 40 inventive principles, such as Mechanics Substitution: Replace a mechanical means with a sensory means (optical, acoustic, or the like), or use electric, magnetic, or electromagnetic fields to interact with the object. This might lead from a diesel locomotive to an electric locomotive or a linear induction motor. Note that the principles do not ensure that the resulting system is the same weight, but they provide a concept for the new system.

Both unstructured and structured methods are necessary in the application of creativity. It is difficult to prescribe unstructured creativity, so it may appear that our view of system architecture more closely reflects structured creativity. Indeed, our contention in the introduction to this text wasis that we would prefer to succeed with an architecture chosen well rather than with one chosen by luck. Simply put, it is our experience that many forms of creativity are at the heart of elegant architectures (see Box 12.1 Principle of Creativity).

We defined the system concept in Chapter 7 as a vision, idea, notion, or mental image that maps function to form. Necessarily, the concept embodies a principle of function and operation and includes an abstraction of form. The architect creates the concept for the system. This is a time of peak creativity, because the ­selection of concept will have a deep and far-reaching impact on the system. The concept should establish the solution-specific vocabulary—it is the beginning of the architecture. The civil architect Steve Imrich notes that “the concept rationalizes the structure of the architecture.” The concept is not a product attribute; it is a mapping from one attribute (function) to another (form).

The concept is separate from the architecture; it is a partial answer. We explicitly separate these two ideas to recognize that concepts are the working language of ideas, whereas architecture is the working language of implementation.

In Chapter 11, we described a method for analyzing concepts in OPM, based on completing the To-By-Using framework with information gathered from the solution-neutral function, the solution-specific function, and form. In this chapter, we move to complex systems where representation in Object Process Methodology may be possible at a first level, but where domain-specific language and methods quickly become more important.

We will structure the concept ideation phase according to the four steps outlined below. In Part 2, we built a framework of questions (Questions 4a to 8b) to first represent the concept and then the architecture. Here we will explicitly focus on generating multiple concepts, essentially an expansion of Questions 5a and 5b of Table 8.1.

1. Develop the Concepts

   • Start with the system problem statement (SPS) and descriptive goals.

   • Analyze (and reinterpret) to identify solution-neutral operands and processes related to value.

   • Apply creativity, and specialize to determine a specific operand/process/instrument concept.

   • Check that each concept meets the system problem statement goals, and reformulate the solution-neutral statement if necessary.

2. Expand the Concepts and Develop the Concept Fragments

   • For rich multifunctional concepts, expand or decompose the concept to reveal ­principal internal function or expanded operands/processes/instruments.

   • For each of these, repeat the steps listed under “Develop the Concept” to identify ­operand/process/instrument concept fragments.

3. Evolve and Refine the Integrated Concepts

   • Search systematically through the space of concepts and fragments to ensure coverage.

   • Combine fragments combinatorially, with constraints, to identify integrated ­ concepts.

4. Select a Few Integrated Concepts for Further Development

   • Apply backwards considerations: What is most likely to satisfy goals?

   • Apply forward considerations: What is most likely to make a good architecture?

   • Check that selected integrated concepts meet SPS and descriptive goals, and reformulate if necessary.

For the remainder of the chapter, we apply this framework to the generation of concepts for developing an architecture for the hybrid gas/electric car.