To Engineer Is Human: The Role of Failure in Successful Design

A deep insight into complex problem solving; it isn't done in a single brilliant step but in smaller steps while learning from our mistakes.

This isn't a very good book in my opinion, and it's a shame because it's a very strong thesis excellently presented. But all too briefly and too thinly demonstrated. A really great idea that seems to have fizzled for some reason.

The basic idea is that engineering is commonly imagined (at least by people outside of engineering) to be a matter of systematically studying a problem and crafting a great solution. The reality, as the author explains very well, is that engineering efforts viewed in retrospect are more like hypotheses or good guesses at solutions. We craft something that works, and then see the weaknesses and learn from them for the next improvement.

The point is that improvements are not simply a matter of meeting new customer demands or adding new features, nor even just correcting avoidable mistakes. Some corrections and improvements are neccessary because complex systems have aspects that really can't reasonably be predicted at design time. The mistakes don't just arise because engineers are less than perfect, but are an intrinsic part of the process of human beings engineering complex designs in the real world. There is rarely if ever even the potential for creating a flawless perfect design that anticipates all likely contingencies and second and third order causal effects of even simple changes.

Whether a "zero defect" mentality is helpful or not as an ideal, it doesn't reflect how engineering actually works, at least when complex systems are involved. The reality of human engineering is that intermediate failures are an important part of the process. Engineering methodology needs to take this into account, and make best use of it, or else lead engineers in a futile struggle for a perfect initial design and forever wonder why they fail.

Although the I think the idea is very solid, useful, and important, this book certainly lacks the depth it needed to make the point and would have made an excellent chapter in a more detailed book. There are important issues raised here, but not answered, about how to improve the engineering process based on this insight into the role of failures.

As a companion to this, I recommend Dietrich Dorner's cognitive science account of the origin of planning failure in complex systems in "The Logic of Failure." Dorner explains in more detail, with the help of his problem situation simulation research, why consequences are so difficult to forsee and plan for.

I gave it one star becuase it wouldnt let me give it less

WOW!!!!! I had to read this for school. That was the worst book I have possibly ever read. He wrote a chapter about cracks in his knives. CRACKS IN HIS KNIVES!!!!!!!!!! Then he told the life story of his son and his slingshot. What kind of crackhead even has the time to write a book in the firstplace, especially one so bad. Mr. Petroski: stick to your day job. How the hell does a book like that get published? Hey if you have spare time, pick up THE PENCIL. Or, you could spend your time watching TV and doing nothing. It would be more value-added than this stupids book. That is unless your having trouble sleeping.

I usually love books that look at the greater meanings in the evolution of engineering/science etc. But this book is completely void of details on why things went wrong. I kept waiting for interesting insights into engineering failures, but all I got was platitudes. And like the other reviewer said, it put me to sleep. If you want to read a book on the role of failure in successful design, look elsewhere.

Don't let the interesting cover fool you. Unless you're a complete nerd, you won't enjoy this book. It will however put you right to sleep within 10 minutes.
Even with an engineering degree, I found this book to be rather dull.

In this enlightening book, Petroski, who is professor of civil engineering, has succeeded admirably in conveying what engineering is and what engineers do in a manner that is accessible even to my grandmother, i.e., the general public. His presentation, although somewhat repetitive, is clear and sprinkled judiciously with humor. Moreover, it is illustrated with familiar analogies, and also numerous mechanical and civil engineering examples including everyday objects such as paper clips, toys and knives.

To engineer is to design, `making something that has not existed before'. Petroski provides insights into the design process (which involves computers extensively nowadays) and its limitations, and also the means employed by engineers to prevent failures in their designs.

He emphasizes, however, that it is not possible to anticipate all possible ways a design can fail and thus failures inevitably occur because engineers are, after all, humans. Numerous examples of catastrophic structural failures throughout history are presented and discussed. All involved the tragic loss of lives (for instance, the collapse of two crowded suspended walkways onto the crowded floor of the Kansas City Hyatt Regency hotel in 1981) except the collapse of the Tacoma Narrows bridge in Washington State in 1940.

Petroksi also discusses the failure analysis or forensic engineering that is performed in the wake of a catastrophic design failure to understand how and why the failure occurred. He argues convincingly throughout the book that understanding such design failures can advance engineering more than successes. Design failures, like other failures in life, should be embraced, rather than denied or ignored, and learned from. Great engineers, and great people in general, are the ones who heed George Santayana's famous dictum: `Those who cannot remember the past are condemned to repeat it.'