This article reviews a concept borrowed from psychology that has given perspectives on product development. Much progress has been made in new areas opened by the systems-level approach to engineering design. Among them are advancements in our understanding of decision-making, ideation, collaboration, modularity, requirements modeling, lifecycle considerations, robust design, green design, and various other "design for x" strategies. The rethinking of design that is needed now is a conceptual basis that allows engineers to better describe and solve problems at the system level, problems that involve user interaction. The concept we propose to deal with these problems is affordance, a term borrowed from perceptual psychology and popularized by the psychologist Donald Norman is his book The Design of Everyday Things. As the theoretical tenets of design are brought up to date with the increasing demands of systems-level design, systems integration, global marketplaces, new materials, new processes, and other recent demands, we anticipate continued growth and advancement in the field of design theory that will benefit all engineers and designers.
Pick an ME design textbook off the shelf, and if it was published before the 1980s or '90s, you're likely to see a heavy focus on analysis methods, literally the nuts and bolts of mechanical engineering. For much of engineering's history, such hard science has dominated design activities. Until recent years, doing design has been synonymous with solving narrowly specified problems, usually at the component level.
Over the past couple of decades, the field of design theory and methodology has been born literally out of the realization that the process of design can be studied, taught, and practiced using scientific methods. Two widely influential texts propelling the field were Herbert Simon's The Sciences of the Artificial and Engineering Design: A Systematic Approach by Pahl and Beitz. An important shift in design thinking advocated by both of these works is that design problems begin at the system level (a whole car, not just a driveshaft). Individual components can be designed only after the whole problem has been understood and defined at the system level and then decomposed into subsystems and so forth, down to individual components that can be designed using the hard engineering sciences and traditional analysis methods. This represents the first major rethinking of design theory since Newtonian mechanics rationalized the design of mechanical components from antiquity.
Programs in New Areas
Much progress has been made in new areas opened by the systems level approach to engineering design. Among them are advancements in our understanding of decision making, ideation, collaboration, modularity, requirements modeling, life cycle considerations, robust design, green design, and various other "design for x" strategies.
Indeed, the scope of disciplines and problems associated with our understanding of design has blossomed to the point that the researcher George Hazelrigg has argued that design is no longer just multidisciplinary, it is omnidisciplinary in that any and all disciplines may be involved in the solution to a particular design problem. Similarly, the late researcher Stuart Pugh advocated "Total Design" wherein every aspect of a design problem must be properly addressed to arrive at a satisfactory solution.
Meanwhile, as the scope of design research and design thinking has expanded, the theoretical foundations that our understanding of design is based on have not been the focus of much attention or progress. There has been little rethinking of engineering design since the seminal works of Simon and Pahl and Beitz. Given the increasing scope and rapid progress in engineering design research, there are those of us who believe it is time to take a step back and examine the foundations upon which our present understanding of engineering design is based. It is time to rethink design theory again.
At Clemson University, we have been looking at an innovative method of design based on a new concept. Interested researchers here include myself as well as several professors, including Georges Fadel, Joshua Summers, Gregory Mocko, and Dina Battisto. We are studying an approach to design that we believe goes beyond the idea of function, and gives a deeper and broader understanding of the issues involved.
Traditionally, it is often stated, or at least implied, that engineering design is all about function. As the popular engineering textbook Product Design states, "All products do something. There is some intended reason behind their existence: the product function." Such an assumption dates back to the approach of Newton and Descartes of ascribing deterministic behavior to machines using mathematics, and it has been extremely successful and reliable. Function, after all, is easily captured mathematically: F(x)=y. When a machine performs a function, the design of that machine is reduced to a math problem.
Things are not so simple in other disciplines that practice design. For example, in the design of consumer products (industrial design) and the design of built environments (architecture), it is not the functionality of the artifacts being designed that dominates. Architects and industrial designers even dare to speak of the emotion and poetry of their designs, terms that (gasp) never stain the pages of typical engineering design texts.
The functional approach is a particularly good fit for the design of traditional mechanical components, such as shafts, bolts, springs, and the like. Indeed, the function based approach of Pahl and Beitz recommends decomposing a problem down to a level in which individual components can be specified to embody individual functions.
However, things get a little more interesting when we broaden our scope to more than just basic mechanical components-that is, when we look at the system level. For example, what is the function of a bridge? Or an automobile? What's more, is the function of a 2008 automobile far different from the function of a 1908 Model T? Is the function of a modern steel and concrete highway bridge any different from the function of the first iron bridge of 1779? The design of automobiles and bridges has changed dramatically while their function has not.
The example of a bridge is particularly interesting because as in all static structures (including furniture, hand tools, artwork, etc.) these artifacts do not do anything. The "intended reason behind their existence" is not to transform any inputs to any outputs. While bridges carry mechanical loads, this is not the essence of their existence. They carry loads for a larger reason. They allow users to do something with them. Bridges allow traffic to pass over a river or gorge. Chairs allow users to sit down. Screwdrivers let users turn screws.
These examples illuminate a problem of description for the function approach. The problem stems from the shift in engineering design from the component level to the system level.
Moreover, problems at this level, including automobiles, bridges, tools, furniture, appliances, and so forth have another feature in common. They all involve user interaction. And as the computer scientist Peter Wegner has pointed out, user interaction is not deterministic and therefore not algorithmic and not functional. All artifacts allow users to do something with them, regardless of how much the artifact does itself.
The rethinking of design that is needed now is a conceptual basis that allows engineers to better describe and solve problems at the system level, problems that involve user interaction. The concept we propose to deal with these problems is affordance, a term borrowed from perceptual psychology and popularized by the psychologist Donald Norman is his book The Design of Everyday Things.
The affordances of a product are what it provides, offers, or furnishes to a user or to another product. Most products have a multitude of such affordances. Automobiles, for example, afford transportation, a comfortable ride, sporty performance, and so forth. Windows afford a view of the outside environment and the admittance of fresh air. These affordances, between artifacts and the people that use them, we call Artifact- User Affordances (AUA). If we consider a bridge, it allows pedestrians to cross a river (an AUA), but it also allows cars driven by people to cross. The latter affordance is between the bridge and the cars, which is an example of what we call an Artifact-Artifact Affordance (AAA).
An important distinction between the affordances of an artifact and its functions is that the affordances depend on the physical form of the artifact, whereas the functions do not. The fact that functions are form-independent is useful in the design process because it frees designers to choose the physical embodiment that best accomplishes each function . However, the fact that affordances are form-dependent can also be useful, because it allows engineers to analyze and compare the affordances of product concepts (especially at the system level) as well as of existing products for reverse engineering.
Positive and Negative Affordances
Another useful feature of the affordance approach is that it is able to describe intended as well as unintended aspects of the product. For example, a motor's function is to transform electrical energy into rotational kinetic energy. Thus, it affords rotary motion, which is desired and is therefore what we classify as a positive affordance. However, because of the various resistances inside the motor, the motor also generates heat, which is undesired and is therefore what we classify as a negative affordance.
By consciously analyzing both the positive and negative affordances of product concepts during the design process, negative affordances can be designed against or mitigated. A similar process is not supported by function-based approaches in which the functions simply capture the "intended" or desired functionality.
A useful tool allowing engineers to use the additional information captured by the affordance approach is the Affordance Structure Matrix (ASM), wherein the system level affordances are mapped to the artifact's individual components. ASMs can be prepared for different artifact concepts or existing products, and then compared to each other or analyzed to attempt to improve various affordances. The interior of the matrix captures which components are related to each affordance. Additional detail can be added by distinguishing whether individual components are either helpful or harmful in achieving a positive affordance or protecting against a negative affordance.
The left side captures which affordances are related to each other, while the top captures which components are related to each other as in a traditional Design Structure Matrix (DSM). The totals on the bottom show how many affordances each component is related to, while the totals on the right show how many components are related to each affordance.
There is also room to describe the quality of the affordances of an artifact. A user can sit on a tree stump just as easily as on a chair. They both possess the affordance of "sit-ability," but this affordance is of higher quality in the chair than in the tree stump. By contrast, the basic function is the same-to support the user's weight. In the case of antique automobiles compared to modern ones, we see that the basic functionality remains largely fixed over time, while the affordances have changed rather dramatically. Original positive affordances have been improved (such as ride comfort), new positive affordances have been introduced (such as those offered by navigation systems), and negative affordances (such as noise and pollution emitted) have been significantly reduced.
The affordance approach also offers an integrative framework in which a larger set of requirements can be captured, including the product functionality, as well as nonfunctional requirements, such as aesthetics, marketability, manufacturability, sustainability, and so forth. Similarly, a wide variety of recently developed design tools can be tied together in the sense that they all address specific affordances of a product.
Since the term "affordance" was coined by the psychologist James Gibson almost 40 years ago, the idea has been applied in a variety of fields, including childhood psychology, the design of graphical user interfaces, mobile robots, control room interfaces, and as a bedrock concept for the field of ecological psychology, which Gibson pioneered in the 1960s and l70s. The application of the concept of affordance to engineering design is a relatively late development.
Augmenting Traditional Approaches
The concept has been attracting increasing attention as other researchers attempt to overcome limitations of traditional engineering approaches, while creating bridges to other design-centric disciplines, such as industrial design, software design, and architecture. Affordancebased design has sparked interest in places as diverse as Worcester Polytechnic Institute, the IIT Institute of Design, the University of Luxembourg, and HyundaiKia Motor Co. I apply it in my own consulting service, Maier Design Works.
An interesting tool developed by Adriano Galvao, a vice president at Sylver Consulting in Chicago, in his book Design Relationships: Integrating User Information Into Product Development is the Function Task Interaction Matrix, which attempts to identify affordances as the intersection of product functionality and user tasks.
We are encouraged that other researchers are also engaged in rethinking some of the fundamentals of design theory. Future frontiers for design theory in this area are the integration of function and affordance-based methods, formalisms for identifying affordances, and vocabularies for naming affordances.
Researchers at Clemson have received National Science Foundation funding for research into affordance-based design. The most recent is for a project led by Gregory Mocko and Georges Fadel, "Integrative Situated Design: Linking Functions and Affordances Through Form," which is getting under way now.
As the theoretical tenets of design are brought up to date with the increasing demands of systems level design, systems integration, global marketplaces, new materials, new processes, and other recent demands, we anticipate continued growth and advancement in the field of design theory that will benefit all engineers and designers.