Skip to Main Content
Have library access? Log in through your library
To Forgive Design

To Forgive Design: Understanding Failure

Henry Petroski
Copyright Date: 2012
Published by: Harvard University Press
Pages: 432
  • Cite this Item
  • Book Info
    To Forgive Design
    Book Description:

    When planes crash, bridges collapse, and automobile gas tanks explode, we are quick to blame poor design. But Petroski, known for his masterly explanations of engineering successes and failures, says we must look beyond design to the interdependency of people and machines within complex socioeconomic systems undreamt of by designers.

    eISBN: 978-0-674-06543-7
    Subjects: History of Science & Technology, Technology

Table of Contents

  1. Front Matter
    (pp. i-vi)
  2. Table of Contents
    (pp. vii-viii)
    (pp. ix-xiv)
  4. ONE By Way of Concrete Examples
    (pp. 1-25)

    When Continental Connection Flight 3407 took off from Newark Liberty Airport for Buffalo Niagara Airport on February 12, 2008, it should have been a short hop in the Bombardier Dash 8 aircraft. However, on that cold winter evening the upstate New York area was experiencing freezing rain, which tends to coat airplane wing surfaces with ice. As it builds up, the ice not only adds weight but also alters the aerodynamic shape of the wing, and this naturally affects the way the plane performs in the air. Because ice buildup can produce dangerous flying conditions, airplane wings are designed with...

  5. TWO Things Happen
    (pp. 26-47)

    It should not surprise us that failures do occur. After all, the structures, machines, and systems of the modern world can be terribly complicated in their design and operation. And the people who conceive, design, build, and interact with these complex things are unquestionably fallible. They sometimes employ faulty logic, inadvertently transpose digits in a numerical calculation, mistakenly overtighten a bolt or undertighten a screw, casually misread a dial, or hurriedly push when they should pull. They also can fail to concentrate, to anticipate, and to communicate at critical moments. At other times, accidents can occur because people cease to...

  6. THREE Designed to Fail
    (pp. 48-80)

    Failure in engineering, as in life, is understandably considered a negative occurrence—something to be avoided at almost any cost. When we hear that a device has broken or a system has failed, we begin to look for causes and culprits even as we are still searching for victims and cleaning up the wreckage. We want to identify how and why each failure occurs so that we can prevent a similar disaster in the future. Was the design at fault? Was the material inferior? Was the construction or manufacture careless? Were the components defective or flawed? Was the maintenance negligent?...

  7. FOUR Mechanics of Failure
    (pp. 81-106)

    When I first encountered the concept of failure in a big way I did not fully grasp its significance, even though an overbearing symbol of it was a looming presence in my daily routine. At the time, I was studying at the University of Illinois, and during my first few years there my home away from home was Talbot Laboratory, a large red-brick building located just inside the city limits of Urbana. The lab, whose western face looks straight across Wright Street to the twin city of Champaign, dates from 1928, when it was dedicated as the Materials Testing Laboratory....

  8. FIVE A Repeating Problem
    (pp. 107-127)

    Recently, while visiting a new dental office to have my teeth checked and cleaned, I was reminded of how much overlap exists in the technical interests of dentists and engineers, especially structural engineers. When the dentist asked me my occupation, I told him that I taught engineering. His response was that dentistry was a kind of oral engineering, to which I nodded and grunted agreement. He also offered that his son almost studied engineering in college, but in the end decided he wanted to be a dentist. This non sequitur reminded me that one of my former students, who wrote...

  9. SIX The Old and the New
    (pp. 128-149)

    I have often been asked, “How long can a bridge last?” In other words, “How long after it is built will a bridge fail?” The answer ranges from negative days to months to decades on the one extreme and from centuries to millennia—and possibly even longer—on the other, depending on such diverse and interrelated factors as design, construction, materials, and maintenance, all of which are affected by the vagaries of economics, politics, corruption, weather, use, and luck. It also depends on what we mean by “fail.” Examples are legion. The Quebec Bridge was not even completed when it...

  10. SEVEN Searching for a Cause
    (pp. 150-174)

    Some structures deteriorate slowly and quietly from within, their progress toward total failure as hidden from view as their buried foundations. But if the flawed system is not discovered in time to do anything precautionary or corrective, we can be taken by surprise by an abrupt and incontrovertible failure, such as the collapse of a bridge carrying rush-hour traffic; the explosion of a space shuttle one minute into a mission on a cold winter morning; or the sudden flooding of a city when its aging levees do not hold. The effect of such occurrences is immediately obvious, but the root...

  11. EIGHT The Obligation of an Engineer
    (pp. 175-198)

    Engineers are always striving for success, but failure is seldom far from their minds. In the case of Canadian engineers, this focus on potentially catastrophic flaws in a design is rooted in a failure that occurred over a century ago. In 1907 a bridge of enormous proportions collapsed while still under construction in Quebec. Planners expected that when completed, the 1,800-foot main span of the cantilever bridge would set a world record for long-span bridges of all types, many of which had come to be realized at a great price. According to one superstition, a bridge would claim one life...

  12. NINE Before, during, and after the Fall
    (pp. 199-222)

    The collapse of the Quebec cantilever span may be Canada’s most famous bridge failure and the inspiration for its Iron Ring tradition, but the most infamous American bridge failure is undoubtedly that of the Tacoma Narrows. When completed in July 1940, the Tacoma Narrows Bridge had the third-longest suspended span in the world, surpassed only by the 3,500-foot main span of the structurally and architecturally revolutionary George Washington Bridge, completed in 1931, and the 4,200-foot central span of the iconic Golden Gate Bridge, which opened in 1937. Unlike those enduring structures, however, the Tacoma Narrows Bridge stood for only four...

  13. TEN Legal Matters
    (pp. 223-240)

    Infamous bridge collapses are not the only devastating accidents to engender debate about their causes. In the immediate wake of virtually any catastrophic failure, especially one that results in loss of life, there is often a cacophony of speculation about exactly what happened and who was to blame. Unless there is a continuing action associated with the disaster, such as an ongoing oil leak in an environmentally sensitive location, the focus soon switches from the failure itself to its cause. Usually before too long, a favored theory emerges, which may or may not be correct but nonetheless becomes the focus...

  14. ELEVEN Back-Seat Designers
    (pp. 241-269)

    In the early days of digital computers, when by today’s standards even large machines had little memory or storage space, many software designers made a decision that at the time appeared to be wise. Whenever a date was input, only the last two digits of the year were used. This not only saved time for data-entry clerks but also conserved space in a computer’s limited storage system and simplified calculations involving things like the interest accrued in a bank account or the payment due on a credit card.

    Abbreviating years was nothing new. Long before the advent of electronic computers,...

  15. TWELVE Houston, You Have a Problem
    (pp. 270-301)

    All things, and especially systems in which people interact with things, fail because they are the products of human endeavor, which means that they are naturally, necessarily, and sometimes notoriously flawed. Some technical flaws may be as innocuous as personality flaws are in the individuals who introduce them, but just as we come to ignore the annoying habits, tics, and traits of people with whom we become familiar and comfortable, so do we tend to ignore the idiosyncrasies of technological parts and systems that we work within and rely upon. Knowing that nothing is perfect, we do not expect perfection...

  16. THIRTEEN Without a Leg to Stand On
    (pp. 302-326)

    Children love trains; they also love cranes. When I was a child, one of my favorite model railroad cars was one that had a crane mounted on it. The crane’s body swiveled, and the two cranks on its side raised and lowered the boom and the hook that hung from it. Derailments, which did not have to be simulated, were frequent, and the crane car was often brought into service to right the toppled rolling stock and lift it back onto the tracks. Cranes were also among my favorite Erector set building projects. My beginner set was limited in parts...

  17. FOURTEEN History and Failure
    (pp. 327-362)

    Everyone, it would seem, should want to draw upon the lessons of the past in order to have a more successful future, but we do not always appreciate where the most valuable lessons are likely to be found. Shortly before President-elect Obama took the oath of office, the five living U.S. presidents—past, present, and then to be—assembled for some historic photos. As they stood before the press in the Oval Office, there was an understandable emphasis on success. Then-President Bush, speaking also for his predecessors, said they would be sharing their experiences with Mr. Obama, and that they...

  18. NOTES
    (pp. 363-394)
    (pp. 395-396)
  20. INDEX
    (pp. 397-410)