Machine Design

Voodoo Engineering

Engineering design lies somewhere between art and science, and unlike other engineering disciplines, is the subject of neverending controversy.

Moshe Alamaro
Visiting Scientist Dept. of Earth,
Atmospheric and Planetary Sciences
Massachusetts Institute of Technology

At the Massachusetts Institute of Technology's Dept. of Mechanical Engineering, for example, the pony-tailed professor, Woodie Flowers, has been telling his design students to read the book Emotional Intelligence, and advising them how to design successfully using the right or the left side of the brain. (I try very hard to use all sides of my brain without much success.)

At the other end of the spectrum is Professor Nam Suh, former head of the Dept. of Mechanical Engineering. He has been trying for the past 20 to 30 years (without much success) to transform engineering design into a science with axioms, laws, and equations, just as in physics or chemistry. Before proceeding with this story I wish to share a relevant anecdote about an attempt to make another academic discipline — political science — more scientific than it really is.

In 1971 Harvard Professor Samuel Huntington — who 20 years later wrote The Clash of Civilizations — authored a paper, "The Change to Change: Modernization, Development, and Politics" in the Comparative Politics Journal. In this paper he used "equations" to describe sociological and political concepts. One of these equations was:

Social mobilization / Economic development = Social frustration

A few years later in 1977 Huntington's paper was used in a political science class taught at Boston University. There, graduate student Ann Koblitz was puzzled by the equation. She reasoned that if A / B = C, then an inverted equation A = B X C must also be true and Huntington's equation could as well read:

Social mobilization = Economic development X Social frustration

What does this equation mean? Ms. Koblitz brought the matter to the attention of her husband, Harvard mathematician Neal Koblitz. He later contributed a chapter in a book entitled Mathematics as Propaganda in which he accused Huntington of "mathematical quackery." Huntington's use of equations, to explain political ideas, wrote Koblitz, produced "mystification and intimidation" as well as "an impression of precision and profundity." Professor Huntington, an eminent geopolitical scientist, has been nominated several times to the National Academy of Sciences. But each time opponents blocked his nomination, citing his use of equations to describe politics. As late as April 2004 Huntington is still not a member of the Academy.

Back to engineering design. Professor Suh has written two books, one in 1990 and the second in 2002, on his "axiomatic design." To eliminate trial and error and guesswork, and to enhance creativity in design, Suh proposes the application of axioms, theorems, and corollaries. To bolster his case, he cites examples from history of how other giants of science used axioms — acceptable truths that cannot be proved or disproved — to arrive at their groundbreaking discoveries. Euclid developed geometry using axioms, Newton explored mechanics, Einstein defined relativity, and now Suh modestly claims that he has, at last, developed the axioms that finally would transform engineering design into a science.

Furthermore, Suh uses equations to explain his axiomatic design. The governing equation is:


where FR = a vector (a fancy word for a list) of functional requirements or attributes of a product such as maneuverability, speed, etc., of an automobile; DP = a list of design parameters or design factors such as steel, paint, electronics, etc.; and DM = a design matrix. Suh's design matrix is not a table with rows and columns like that found in management texts. Rather it is a mathematical matrix borrowed from a branch of mathematics called linear algebra. The axioms, theorems, and corollaries are a set of rules that define the proper mathematical use of the design matrix and the lists of a product's attributes and factors.

When I heard about Suh's axiomatic design equation, my first reaction was similar to that of Ann Koblitz. Can Suh's equation be reversed? Can we desynthesize a product and find mathematically its design factors and processes? According to Suh's 1990 book The Principles of Design, the answer is "yes." This means that whether one is analyzing a Rolls Royce or a Ford Escort, for example, it is possible to desynthesize each car into its basic components; indistinguishable piles of steel, plastic, and electronic scrap. But where in the inverted axiomatic design equation is the mathematical element that makes the Rolls Royce and the Ford Escort so different? I reason, using basic principles of linear algebra, that if the answer does not appear in the inverted equation, it also does not exist in the original design equation. Therefore there is no way to design mathematically a Rolls Royce or a Ford Escort. Oh yes, I forgot. Suh's approach does distinguish between the two piles of scrap: The Rolls Royce's pile is bigger.

In his books, Suh reviews past inventions such as the steam engine, invented by Newcomen (later modified by Watt) and shows how they could be developed using his axiomatic design. Impressive, to be sure, but could Suh teach a contemporary designer or inventor such as Dean Kamen, the creator of the Segway personal transporter and many other important medical devices, how to invent better using axiomatic design? Definitely not. Kamen did not or barely finished college and, as a self-taught physicist, probably doesn't know a whole lot about linear algebra and matrix manipulation necessary for axiomatic design. Is axiomatic design a solid science or nothing but mathematical quackery that gives a false impression of precision and profundity? You decide.

Incidentally, Dean Kaman was elected to the National Academy of Engineering in 1997. As late as April 2004 Suh is still not a member of the Academy.

Those who wish to participate in a Web forum discussion on the excessive mathematization of engineering design please visit:

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.