W 1 Sr^rT" Ta.

u>(
t)
KOe<0/ Billy V&U.tf\ (^003) DISCUSSION OF THE METHOD
Conducting the Engineer's Approach to
Problem Solving
pfe

Qj!Lp I - Seme i ^ ^ u ^ / r

Billy Vaughn Koen

New York Oxford

OXFORD UNIVERSITY PRESS
2003
 For Deanne, Kent, and Doug

Oxford University Press

Oxford New York

Auckland Bangkok Buenos Aires Cape Town
Chennai Dar es Salaam Delhi Hong Kong Istanbul Karachi
Kolkata Kuala L u m p u r Madrid Melbourne Mexico City Mumbai Nairobi
Sao Paulo Shanghai Taipei Tokyo Toronto

C o p y r i g h t © 2 0 0 3 b y Billy V a u g h n K o e n .

Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York, 10016
http://www.oup-usa.org

Oxford is a registered trademark of Oxford University Press

All rights reserved. N o part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data

Koen, B. V.
Discussion of the method : conducting the engineer's approach to problem solving
/Billy Vaughn Koen.
p. cm.
Includes index.
ISBN 0-19-515599-8.
1. Methods engineering. 2. Problem solving. I. Title

T60.6 .K(x>5 2003

620—dc21
2002193018

Printing number: 9 8 7 6 5 4 3 2 1

Printed in the United States of America

in\ acid-free paper
Some Thoughts on Engineering

T he use of the engineering m e t h o d rather than the use of reason is hu-

m a n k i n d ' s most equitably divided e n d o w m e n t . I mean by

the engineering method the strategy for causing the best change in a poorly un-
derstood situation within the available resources

and by reason, the ability to distinguish between the true and the false, or what
Descartes has called "good sense." Whereas reason had to await early Greek
philosophv for its development, is even n o w denied in some cultures, and is in
retreat in others, the underlying strategy that defines the engineering method
has not changed since the birth of h u m a n s . To be human is to be an engineer.
The objective of this chapter is to p r e p a r e the way for a consideration of the
strategy the engineer uses to solve problems that will be given in Part II. To
achieve this objective we will divide our thoughts into three sections. In the first,
we will view the engineer through the eyes of the layperson, historian, etymol-
ogist, lexicographer, and anthropologist to learn h o w w e are to recognize an en-
gineer w h e n w e meet one. In the second, attention shifts to the characteristics
of a problem that requires the talents of this new acquaintance. Finally, in the
third part of this chapter, w e will examine several problems that unexpectedly
bear a resemblance to engineering problems. We should anticipate that they
might be effectively attacked using the engineering m e t h o d . Incidentally we will
come to appreciate how long ago it was that ancient man* first encountered en-
gineering problem situations and h o w frequently m o d e r n man confronts them
today. Acceptance of the claim of identity between h u m a n and engineer will
come subliminally as the engineering method becomes more fully understood.
Consideration of the correct epistemological status of reason, the engineering

T h e rules of contemporary writing in America require the author to go through

wild contortions to avoid the sexist use of masculine nouns and pronouns. This
was not the state of the art in 1982 when the first version was completed. With an
apology to those on both sides of this question, we will retain the original format
that referred to the engineer as male. At least initially, the pronoun she will refer
to everyone else: philosophers, scientists, artists, poets, and so forth. This serves as
an excellent example of the changing rules of thumb in acceptable practice over
time.
 Some Thoughts on Engineering / 9
8 / DISCUSSION OI ; THE METHOD
pie "engineering a bill through congress." But despite these frequent examples
m e t h o d ' s chief rival as a basis of the universal m e t h o d , will be reserved until a
to the contrary, w h e n a definition of engineering is sought, the usual tendency
m u c h later chapter in this discussion. is to look to concrete objects such as chemicals, airplanes, bridges, or roads in-
stead of to the method that brought each of these engineering devices into
existence.
This same confusion between art and artifact exists in efforts to date the birth
THE ENGINEER
of h u m a n k i n d . Most anthropologists define the h u m a n by his use of tools, none
Most people think of the engineer in terms of his artifacts instead of his art. As as eloquently as Loren Eiseley in his book The Firmament of Time:
a result they see diversity w h e r e they should see unity a n d find it h a r d to ac-
cept the identity of m a n and engineer. The question, " W h a t is an engineer?" is Massive flint-hardened hands had shaped a sepulcher and placed flat stones
usually a n s w e r e d by "a person w h o m a k e s chemicals, airplanes, bridges, or to guard the dead man's head. A haunch of meat had been left to aid the dead
roads." From the chemicals, the layperson infers the chemical engineer; from the man's journey. Worked flints, a little treasure of the human dawn had been
airplanes, the aeronautical engineer; and from the b r i d g e s a n d roads, the civil poured lovingly into the grave. And down the untold centuries the message
engineer. N o t only the layperson, but also the engineer, m a k e s this mistake. Be- had come without words: "We too were human, we too suffered, we too be-
cause the connection of the engineer with his completed design is so e n d u r i n g lieved that the grave is not the end. We too, whose faces affright you now,
knew human agony and human love."
a n d the connection with his use of m e t h o d so fleeting, a p e r s o n insists he is an
engineer based on w h a t he p r o d u c e s , irrespective of h o w he goes about it, in-
stead of insisting that he is an engineer based on how he goes about it, irre- In this miniature drama w e see a problem born of the most h u m a n craving
spective of w h a t he produces. In a similar fashion, the historian uses the exis- to create an appropriate resting place for a loved one and a solution pitifully
tence of d a m s on the Nile, irrigation canals in various parts ol the ancient world, constrained by severe limits in knowledge and resources. Recalling the defini-
g u n p o w d e r , and pottery to infer the existence of engineers a n d craftspersons in tion of engineering method as the strategy for causing the best change in a poorly
past civilizations. But behind each chemical, each road, each pot hides the com- u n d e r s t o o d situation within the available resources, can w e d o u b t that at least
m o n activity that b r o u g h t it into being. It is to this unity of m e t h o d that w e m u s t as early as Neanderthal m a n w e find engineers?
look to see the engineer in every m a n . In dating the birth of h u m a n k i n d , other scholars a r g u e that it is more im-
To be sure, w e occasionally see an e m p h a s i s on m e t h o d instead of object, portant that "primitive m a n stuck feathers in his hair" than that he worked in
ironically most frequently in the writings of the nonengineer. We read in the flint. We are told that our error in defining h u m a n s by their use of tools results
S u n d a y n e w s p a p e r s u p p l e m e n t that a reigning prince engineered the divorce of from society's present preoccupation with materialism and technology on the
his d a u g h t e r , in the daily n e w s p a p e r that the clergy in Iran engineered the firing one hand and the lack of traces of early myths, customs, and literature in the
of the president, in a medical report that doctors h a v e engineered a better bacte- physical record on the other. "If only w e could still see the pictures of earliest
rial host, and in a book on the g a m e of chess that white has engineered a perfect man and hear him sing," the humanist laments, " w e would define him by his
counter to black's opening. Even the headline of one of Dear Abby's n e w s p a p e r arts instead of his tools." Again the emphasis is on an object such as a glyph or
columns offers the advice, "Don't engineer a problem; just keep silent about past." a song, but in this case it is just a more ephemeral one.
Similar statements using the w o r d engineering in the sense of creating a desir- Each of these views is too n a r r o w and self-serving to be a proper definition
able change in an uncertain situation within the available resources are found of h u m a n k i n d . For behind the earliest crude flint, behind the earliest artistic
daily in novels, reviews, and n e w s p a p e r s and heard on radio and television. scrawl lies a c o m m o n process that brought it into being. We d o not see this
In each of these cases w e sense that the w o r d is being correctly used, a n d w e method because its earliest use, like the earliest picture or song, left no traces.
But a c o m m o n sense of method—of desirable change in an u n k n o w n , resource-
are right. According to one of E n g l a n d ' s most noted nineteenth-century engi-
limited w o r l d — p e r v a d e s and certainly predates both the tools we find and the
neers, Sir William Fairbairn, quoted in Technology and Change (ed. by Burke and
arts we infer. T h r o u g h o u t history, differences m a y have been observed in how
Eakin), well h u m a n s have used the engineering method or in the subject matter they
have treated, but the use of the engineering m e t h o d is coterminous with any
The term engineer comes more directly from an old French word in the form
reasonable definition of the h u m a n species. The engineering method, like a midwife,
of a verb—s'ingenier. . . and thus we arrive at the interesting and certainly lit-
was present at the birth of humankind.
tle known fact, that an engineer is . .. anyone who seeks in his mind, who sets
his mental powers in action, in order to discover or devise some means of suc-
ceeding in a difficult task he may have to perform.
CHARACTERISTICS OF AN ENGINEERING PROBLEf
The dictionary concurs by authorizing the verb to engineer as "to contrive or
plan usually with more or less subtle skill or craft" and by giving as an exam-
The engineer's method itself, not the results of this method, legitimates the word
ple "to engineer a d a r i n g jailbreak." It also gives a second definition: "to guide,
engineer in all of the previous examples, and it is by this method that the engi-
m a n a g e , or supervise d u r i n g production or d e v e l o p m e n t , " including the exam-
 Some Thoughts on Engineering / 11
10 / DISCUSSION OF THE METHOD
lem. Again the Golden Gate Bridge will serve as a good example. Design of a
bridge is a complex operation requiring knowledge of strength of materials, wind
conditions, economics, politics, mathematics, fluid flow, and so forth. No engi-
neer or team of engineers could ever h o p e to acquire all of the information nec-
essary to design the perfect Golden Gate Bridge, even if such an absolute ideal
were to exist. This uncertainty, this complexity, this d o u b t about the system or
about the exact problem statement itself—this poorly u n d e r s t o o d situation—is
the sure mark of a problem that requires an engineer to roll u p his sleeves and
go to work.
N o w we will look in detail at the key words change, resources, best, and un-
Time certainty, all of which have a p p e a r e d in the definition of an engineering prob-
lem situation. Of these four, the reasons for including the first two are relatively
Figure 1 Measure of Change
easy to explain. The third is less well understood a n d more philosophically
charged. It will occupy more of our time. T h r o u g h o u t the discussion of these
first three key words we will sense the fourth, the lack of information or the un-
certainty that always pervades an engineering problem, menacing us from the
neer should b e recognized. W h e t h e r w e consider h u m a n s at their birth, Nean-
wings. After this mise-en-scene, the engineering strategy itself can m a k e an ap-
derthals, the ancient d a m builders, the prince w h o w a n t s the divorce of his
pearance in Fart II of this discussion.
d a u g h t e r , or the designer of a m o d e r n bridge, each is correctly called an engi-
neer because each finds himself in an equivalent kind of situation and reacts in
a similar w a y . Each w a n t s to change, to modify, or to convert the world repre- Change
sented by o n e state into a w o r l d represented by a different one. The initial state
Engineers cause change. I know of n o engineer w h o would disagree with that
m i g h t be San Francisco w i t h o u t the Golden Gate Bridge; the final state, San Fran-
statement. Prompted by their senses, most people will agree with the engineer
cisco with this bridge. The initial state might be the Nile w i t h o u t a d a m ; the fi-
that change does occur and that the world is s o m e h o w different from one mo-
nal one, the Nile with a n e w d a m in place. Or the initial state might be a N e -
ment to the next. Close the book you have in your h a n d and reopen it. Can you
a n d e r t h a l c o n t e m p l a t i n g the d e a t h of a loved one; the final one, the world after
d e n y that it was open, then closed, a n d n o w open once again? The engineer, if
the construction of a sepulcher. But w e can also view the m a n e u v e r i n g to ob-
he thinks about it at all, finds it hard to believe that a n y o n e could seriously ques-
tain a divorce, the firing of the president of Iran, or the creation of a better bac-
tion the existence of change, especially the change he causes, and dismisses any-
terial host in a completely a n a l o g o u s way. Can w e deny that the world is some-
one w h o prolesses to do so as a charlatan. An engineering colleague once told
h o w different after the composition of a song, the painting of a picture, or the
one of my classes, "I consider a n y o n e w h o is not actively engaged in causing
construction of a rhyme? Graphically each of these examples is represented in engineering change a leech on society." Fortunately for him, most Western
Figure 1, w h e r e time is given o n the horizontal line or axis and s o m e m e a s u r e philosophers would at least agree that change can occur, but there have been
of change in the w o r l d is given on the vertical axis. some notable exceptions. In ancient times Parmenides taught that change was
The e n g i n e e r is to cause the transition from A to B. To identify a situation re- impossible and felt that the changes w e seem to perceive are b u t illusions of the
quiring an engineer, seek first a situation calling for change. senses. Surely it must seem to us in the modern age that Parmenides was delu-
Yet from an engineering point of view not all changes or all final states are sional, and his claim u n w o r t h y of being used as an example. Is there any chance
equally desirable. Today few w o u l d suggest replacing the Golden Gate Bridge that the delusion is instead ours? More recently Santayana has held that doubt
as a m e a n s of crossing San Francisco Bay with one of the w o o d e n covered bridges is always possible about the existence of actual change, and McTaggart has ex-
that w e r e once c o m m o n l y seen in Maine. To identify a situation calling for the plicitly denied that change exists. The engineer m a y be startled to learn that
engineer, second w e m u s t look for one in which not just any change, b u t the Leibnitz, w h o in parallel with N e w t o n invented the branch of mathematics most
best c h a n g e , is desired. concerned with change—the calculus—had serious reservations about the mat-
Unfortunately the engineer cannot select this best change from all conceiv- ter. In addition, consideration of the work of Bradley and expansion of our dis-
able transitions from the initial state to the final one. Physical, economic, polit- cussion to include the view of Eastern philosophy, w h e r e denial of change is
ical, a n d artistic constraints a l w a y s exist. Despite its favorable corrosion p r o p - even more prevalent, s h o w that, contrary to the belief of my engineering col-
erties n o consideration w a s given to building the Golden Gate Bridge of an alloy league, there are serious philosophical problems in determining if change exists
of p u r e gold, for obvious reasons. In the third place, then, the engineer a l w a y s and what it is. By insisting that an engineer is involved in changing the world,
seeks the best change within the available resources. we have taken a philosophically risky first step. Given the equally ambiguous
The final characteristic of a situation requiring the engineer is that k n o w l e d g e notion of causality in m o d e r n philosophy, by professing that engineers can cause
about the system before, d u r i n g , and after the transition is incomplete, incon- this change, we have taken an even more precarious second one. just for the
sistent, or w o u l d require m o r e time to accumulate than the lifetime of the prob-
 Some Thoughts on Engineering / 13
12 / DISCUSSION OF- THE METHOD

m o m e n t , note well, w e will a s s u m e that change occurs and will consider that Resources
change for w h i c h the engineer claims responsibility. The second characteristic of a situation that requires the services of an engineer
We i m m e d i a t e l y r u n into four practical difficulties w h e n w e consider the is that the desired solution m u s t be consistent with the available resources. These
changes for w h i c h the engineer claims credit: The engineer doesn't k n o w w h e r e resources are an integral part of the problem statement and both define and con-
he is, w h e r e he is going, h o w he is going to get there, or if a n y o n e will care w h e n strain its solution. Different resources imply different problems, and different
he does. Initially, the engineer is located at point A in Figure 1. No one w o u l d problems require different solution techniques. To make this point, one of my
claim to h a v e perfect k n o w l e d g e about the w o r l d at the time represented by this former engineering professors* would begin each class period with a simple
point. T h e second p r o b l e m is that the exact final state, point 15, is also u n k n o w n problem to be answered in fifty seconds by what an engineer would call a back
w h e n the e n g i n e e r first goes to work. An example will m a k e this point clear. of the envelope calculation. Once, for example, we were asked to estimate the n u m -
The A s w a n H i g h D a m in Egypt h a s increased the salinity of the Nile by 10 per- ber of Ping-Pong balls that would fit into the classroom. In addition to devel-
cent, h a s led to the collapse of the sardine i n d u s t r y in the Delta, has caused oping the ability to manipulate large n u m b e r s in our heads, these problems
coastal erosion, a n d has forced the one h u n d r e d t h o u s a n d N u b i a n s displaced taught the importance of resources in the definition of a project. Based on fifty
by the reservoir to try to a d a p t to life as farmers on the newly created arable seconds w e were to provide an a n s w e r to the problem—a correct engineering
land. These liabilities h a v e been balanced, s o m e w o u l d say more than offset, by answer. H a d w e been given two d a y s for our response, we would have been ex-
other assets s u c h as the generation of e n o u g h hydroelectric p o w e r to furnish pected to measure the room a n d calculate the number—again, an entirely cor-
one-half of E g y p t ' s electrical n e e d s , etc. O u r interest, however, is not to critique rect engineering answer given the additional resources. I suppose if w e were
this spectacular engineering project or to reconcile conflicting opinions as to its given even more time and needed the best possible answer, we should have
net w o r t h , b u t to e m p h a s i z e that before construction, at state A, the engineer filled the room with Ping-Pong balls and counted them. Although obviously
could not h a v e predicted the exact change in salinity and erosion or the exact similar, each of these problems was fundamentally different as evidenced by
h u m a n costs to the s a r d i n e fishermen a n d the N u b i a n s . The final state always their need for diflerent m e t h o d s of solution. The a n s w e r to each was absolutely
has a reality that an engineer situated at the initial state cannot anticipate. Like- correct from an engineering point of view w h e n both the problem and time con-
wise, the presidential o r d e r to " p u t a m a n on the m o o n by the end of the d e c a d e " straint were considered together.
lacks the specificity of the l a d d e r Neil A r m s t r o n g descended to leave his foot-
print on the m o o n . The engineer is willing to d e v e l o p a transition strategy from Contrast an engineering problem to a scientific one with respect to their de-
the p r e s e n t to a p r e s u m a b l y m o r e desirable future state, but rarely is he given p e n d e n c y on resources. Although N e w t o n w a s limited in the a m o u n t of time he
a specific, well-defined p r o b l e m to solve. Instead he m u s t determine for himself had to develop his theory of gravitation a n d a m o d e r n cancer researcher is con-
w h a t the actual problem is on the basis of society's diffused desire for change. strained by the funds at her disposal, we usually think of each as trying to read
At the b e g i n n i n g of an engineering project the engineer rarely k n o w s exactly the already written book of N a t u r e instead of creating a new best-seller based
upon the available resources. We quibble, by extending the analogy beyond its
w h e r e he is going. bounds, if we try to argue that Nature, and by implication science, has a correct
The third difficulty is with the change itself. Figure 1 falsifies the ease in de- answer to the Ping-Pong ball problem and that the engineer is only limited bv
ciding w h i c h p a t h to take from A to B by s h o w i n g only one. Usually a n u m b e r his resources in his approximation to this n u m b e r . A similar sense of conver-
of alternatives exist, each limited by different constraints. The engineer is not re- gence to truth does not usually exist in an actual engineering problem. For ex-
sponsible for i m p l e m e n t i n g a single given change, b u t for choosing the most ap- ample, if we try to a r g u e that N a t u r e has an absolutely correct a n s w e r as to
propriate o n e . In other w o r d s , at state A he d o e s n ' t know h o w he is going to whether the Aswan High D a m should have been built and that the engineer will
get to state B even if he w e r e to h a v e complete k n o w l e d g e of the p r o b l e m he is converge to it with additional resources, w e quickly become inundated in pro-
to solve. found philosophical water. Instead of looking for the answer to a problem, as
The final difficulty in causing change is that an engineering goal has a way does the scientist, the engineer seeks an answer to a problem consistent with the
of c h a n g i n g t h r o u g h o u t a design. From the start of a project to completion is of- resources available to him. This distinction will become clearer when we later
ten a long time. At present, for example, it takes twelve years to construct a nu- consider the engineer's notion of best. At least for now it seems reasonable to
clear reactor in America. D u r i n g the completion of an engineering project, agree with the famous engineer, Theodore Von Kantian: "Scientists discover the
changes in t h e final goal often occur requiring a reorientation of the project in world that exists; engineers create the world that never was."
midstream. In the automobile industry the public's d e m a n d has flitted from the An engineering problem does, however, require evaluation of the true re-
desire for a powerful automobile, to a safe automobile, to a small, fuel-efficient sources that are available. This is difficult for three reasons. We usually think of
one, a n d back to a larger one at a rate that h a s often left a new automobile de- resources as time and m o n e y because they are often explicitly stated in the prob-
sign obsolete before it leaves the d r a w i n g board. With the lack of information lem, are usually in short supply, and are depletable. In exceptional cases other
about points A and B a n d the desired transition p a t h between them, combined depletable resources may also be important. Since a good engineering rule of
with c h a n g e s in point B t h r o u g h o u t the project, h o w can the engineer ever h o p e thumb is that approximately one h u n d r e d p o u n d s of rocket are required to place
to cause the c h a n g e he desires? C h a n g e is recognized as a characteristic of an
engineering p r o b l e m , b u t w i t h all of the a t t e n d a n t uncertainty, w h a t strategy *Ur. |. |. McKetta, Jr.
does the engineer use to achieve ID
14 / DISCUSSION OF THE METHOD Some Thoughts on Engineering / 15

one pounci of satellite into orbit, weight is often limited in space exploration, sumnbly overall productivity of the work force. When the flexibility of allocat-
and the engineer treats it as a precious resource to be wisely m a n a g e d . ing resources and exchanging one for another is decreased, the design often suf-
But depletable quantities such as time, money, and weight are only one part fers. In the project to p u t a m a n on the m o o n , the deadlines within the project
of w h a t is m e a n t by a resource in an engineering context. An engineering re- were set, in large measure, by political considerations, and, hence, were rela-
source is a n y t h i n g that m a k e s an i m p o r t a n t or significant difference to a proj- tively inflexible. As a result preliminary suborbital and orbital missions were
ect. To d e t e r m i n e if s o m e t h i n g should be considered as a resource, perform this scheduled so closely together that time did not permit the results from one ex-
simple test: Imagine two t e a m s of engineers identical in all respects but one. If periment to be analyzed and used in the design of the following one. Instead,
the final p r o d u c t of one team is judged preferable to the p r o d u c t of the other, results from one experiment could only affect later missions in a leapfrog man-
then the difference b e t w e e n the two teams should be taken as a resource. As an ner. Putting a man on the moon in a long period of time as opposed to "by the
example, consider two engineering teams with the same n u m b e r of members, end of the decade," although unacceptable politically, would have been far less
costly. The efficiency with which resources are used is as important a consider-
same education, and same a m o u n t s of time and m o n e y at their disposal. O n e
ation as the total a m o u n t in defining and limiting an engineering project.
team, h o w e v e r , is m o r e experienced solving p r o b l e m s similar to the one w e pose.
Past experience with similar p r o b l e m s usually p r o d u c e s a better design. Past ex- An engineering problem is defined and limited by its resources, but the true
perience with similar problems should, therefore, be considered as a resource, al- resources must be considered. Because w e tend to think only in terms of de-
t h o u g h obviously it is not a depletable one. As a second example, consider the pletable resources, because we confuse nominal and actual resources, and be-
project of writing a book. Money and time to aid in the e n d e a v o r are clearly re- cause we neglect the efficiency of allocating resources and the possibility of ex-
sources, b u t w h a t of interest and encouragement? I believe the author given these changing one kind for another, often the true resources are hard to determine.
last t w o effectively has been given a resource, and both the speed of completion Given the uncertainty in determining what should be considered a resource,
and the quality of the writing are enhanced. Again, the important point is that how can the engineer ever hope to provide an a d e q u a t e solution to a problem?
neither interest nor encouragement is depletable. By a similar analysis, team com- What strategy does he use?
patibility, organizational ability, a n d enthusiasm should also be considered as im-
portant resources. The first difficulty in d e t e r m i n i n g w h a t should count as a re- Best
source is focusing exclusively on depletable resources.
A second difficulty is confusion b e t w e e n the nominal resources (those ex- We have agreed to consider the four characteristics of an engineering problem
plicitly n a m e d in the p r o b l e m statement) a n d the actual resources. Perhaps a captured in the k e y w o r d s change, resources, best, and uncertainty. The first two of
problem is nominally s u p p o s e d to be finished within a certain b u d g e t or within these w e r e easy to u n d e r s t a n d . We can easily accept that an engineer wants
a fixed time period, b u t c o m m o n practice allows a 10 percent overrun in either. change and, with s o m e thought, almost as easily accept that the definition of an
T h e design proposal that is u n a w a r e of this fact is inaccurate a n d , in competi- engineering problem d e p e n d s on the available resources. The third characteris-
tive b i d d i n g , at a definite d i s a d v a n t a g e . Large g o v e r n m e n t projects are espe- tic of a problem that requires the services of an engineer, the appropriate notion
cially p r o n e to this effect. Another example is the m a n a g e m e n t of funds for re- of best, is more of a challenge to understand and m u s t occupy m o r e of our time.
search into a serious disease, such as cancer. Instead of using all of the available The engineer w a n t s the best solution to his problem, but then w h o wouldn't?
m o n e y on research, c o m m o n practice is to use some of the m o n e y for a fund- The difficulty is that the engineer's notion of best represents a surprising de-
parture from the s t a n d a r d concept as e n d o r s e d by most Western philosophers
raising drive. In this w a y the actual resources for research m a y be m a n y times
and as understood by most nonengineers. Most of us would accept Plato's no-
greater than the original n o m i n a l ones.
tion (without recognizing to w h o m w e o w e d the debt) of an ideal, perfect form
T h e final error m a d e in evaluating the resources available to a project is ne-
of, say, beauty, justice, or whatever as an ultimate best and then consider ap-
glecting the efficiency of their use. A l t h o u g h the total nominal resources are
proximations to this form as better and better as they approach this ideal. In
fixed, allocation within this total is m o r e flexible. Given two design teams, the
fact, it is even difficult for the philosopher to conceive of an alternative to Plato's
one that most strategically allocates its resources or converts one kind into an-
linear progression toward the ideal form if she has not studied engineering. Since
other at the a p p r o p r i a t e m o m e n t has effectively increased the a m o u n t available.
our ultimate goal is to discover a universal m e t h o d by generalizing the method
Money may be used to assemble a more experienced design team or to b u y more
used by the engineer, we must have a very clear u n d e r s t a n d i n g of what the en-
e q u i p m e n t . Time m a y be allocated to search the literature for relevant informa-
gineer m e a n s by his terms before we proceed. This is especially critical when
tion or to increase the basic research to obtain this information. A good alloca-
the engineer's definitions are at o d d s with both c o m m o n a n d accepted philo-
tion strategy effectively increases the available resources. In addition, one re-
sophical practice. W h a t w e n o w seek, therefore, is the engineer's notion of a best
source m a y be converted into another. For instance, hiring additional people at
solution or w h a t is technically called the optimum solution. What w e will find is
certain steps in a project effectively exchanges m o n e y for personnel. But if com-
a new, radical concept of best little used in Western, Greek-based philosophy.
pletion of this step is critical, in the sense that failure to complete it blocks fur-
This a d v e n t u r e will carry us into the heart of an important area of engineer-
ther progress a n d forces other people a n d machinery to remain idle, the in-
ing called optimization theory, which is familiar, at least in outline, to all engi-
creased expense on personnel will result in an overall saving. The coffee break
neers but relatively unheard of and more challenging to the philosopher and
is a curious example of the exchange rate between resources. Here a small
nonengineer.
a m o u n t of time is p u r p o s e l y given u p for an increase in enthusiasm and pre-
 Some Thoughts on Engineering / 17
16 DISCUSSION OF THE METHOD

This investigation will be w o r t h w h i l e because, in addition to defining the en-

gineer's notion of best, it will indicate, once again, the crucial lack of informa-
tion that is a l w a y s present w h e n an engineer solves a problem a n d w i t h which
his m e t h o d m u s t cope. After s o m e preliminaries, the discussion will be divided
into three parts: a simple example to explain the most i m p o r t a n t technical terms
used in optimization theory, a short theoretical analysis, and finally s o m e prac-
tical considerations. Mentioning optimization theory is not m e a n t to imply that
till engineers use it formally in all projects. Instead it is included at this point to
capture the essence of the engineer's notion of best.
Best is an adjective applied r e d u n d a n t l y to describe an existing engineering
design. That a specific automobile exists proves that it is s o m e engineer's sub-
jective notion of the best solution to the p r o b l e m he w a s given to solve. Saying 4 5 6 7
that a Mercedes is a better automobile than a M u s t a n g is nonsensical if better is Knob Setting
being used in an engineering sense. They are both o p t i m u m solutions to differ- Figure 2 Sharpness vs. Knob Setting
ent specific design projects. Likewise, the complaint that "American engineers
cannot build an a u t o m o b i l e that will last for fifty y e a r s " can only be voiced by
a person with little u n d e r s t a n d i n g of engineering. To construct such an auto- s p o n d s to a sharpness of 8. The sharpness of the picture is higher for a setting
mobile is well within the ability of m o d e r n a u t o m o t i v e engineers, but to d o so of 6 than for one of 2. For no compelling reason this curve is often called a re-
is a different design p r o b l e m from the one currently given to the American en- turn function.
gineer. It does m a k e sense to prefer one design project over the other. An engi- A different return function, Figure 3, exists for the s o u n d . Turning the knob
neer could conceivably a r g u e that designing an automobile similar to the Mer- to a higher setting decreases the fidelity of the sound. W h e n the control knob is
cedes is a better goal than designing one similar to the M u s t a n g because it w o u l d set to the value 4, the value of the fidelity is also 4, as can be seen on the graph.
last longer, conserve natural resources, p r o m o t e national pride, or whatever. To represent pictorially w h a t a person does instinctively w h e n he selects his
A n d , of course, a second engineer m a y feel that he could have p r o d u c e d a bet- preferred setting, the two return functions m u s t be combined or superimposed
ter final p r o d u c t than the first engineer given the same p r o b l e m statement. But on the same graph; they m u s t be p u t on the same basis; or, technically, they
for the engineer w h o designed the M u s t a n g , the automobile you see before you must be m a d e commensurate. This requires that a common measure of goodness
is his best solution to the problem he w a s given to solve. To exist is to be some be found for picture and sound. W h a t the engineer seeks is the relative impor-
engineer's notion of best. tance of sound and picture to the o w n e r of the television set. In other words, a
Defining the technical terms used in optimization theory will help to m a k e 10 percent increase in the sharpness of the picture is worth w h a t percentage de-
this point, a l t h o u g h the discussion that follows m a y be s o m e w h a t tedious for crease in the fidelity of the s o u n d ? This relative importance is expressed by the
the nonengineer. Still, it should be carefully studied for it has p r o f o u n d philo- weighting coefficients of the two conflicting criteria. Another term used for the set
sophical consequences for the definition of universal m e t h o d . of weighting coefficients is the value system of the problem.
Consider a television set with one control. We will a s s u m e that turning this Let us assume for the m o m e n t that sharpness a n d fidelity are equally desir-
knob to a higher n u m b e r will p r o d u c e a better picture b u t at the s a m e time able. That is, the relative weights in the two cases are equal. The resulting corn-
worsen the s o u n d ; turning the control to a lower n u m b e r , on the other h a n d ,
will w o r s e n the picture b u t i m p r o v e the s o u n d . Confronted w i t h such a device,
it w o u l d be relatively simple for you to adjust it for y o u r personal preference as
you balance the relative importance of picture and s o u n d to you. The engineer's
job, h o w e v e r , is not only to please you, b u t also to find the best p e r m a n e n t ad-
justment of this k n o b to please society. Let us see h o w this is d o n e theoretically.
The setting of the k n o b is called a manipulated variable. O n e can set it to a lower
or higher n u m b e r as desired. The quality of the picture (say, sharpness) a n d the
quality of the s o u n d (say, fidelity) are the criteria in the p r o b l e m . It is against
these two characteristics that a j u d g m e n t is m a d e as to w h e t h e r or not the con-
trol setting is best. In general, the criteria are conflicting in that an i m p r o v e m e n t
in one implies a w o r s e n i n g of the other. The criteria taken together m a k e u p the
optimization space or axis system of the problem. Figure 2 is a g r a p h that s h o w s
h o w the s h a r p n e s s of the picture might change with the position of the knob. 3 4 5 6 7
As the n u m b e r on the horizontal axis increases, the sharpness, s h o w n on the Knob Setting
vertical axis, is i m p r o v e d . The dotted line indicates that a setting of 4 corre- Figure 3 Fidelity vs. Knob Setting
 Soine Thoughts on Engineering / 19
18 / DISCUSSION OF THE METHOD
udices. If a nation feels that a funeral pyre should be aligned in a north-south
direction to aid the d e a d ' s journey into heaven, the model to be optimized will
incorporate this consideration as a design criterion irrespective of the truth of
the claim. Likewise, the engineering model is not based on an eternal or absolute
value system, but on the one thought to represent a specific society. In a soci-
ety of cannibals, the engineer will try to design the most efficient kettle. As a re-
sult, the o p t i m u m obtained from this model does not pretend to be the absolute
best in the sense of Plato, but only the best relative to the society to which it ap-
plies. Contrast tins with a scientilic model. Speaking of Einstein's theory as the
best available analysis of time and space implies that it comes closest to de-
scribing reality than any alternative proposal. It is considered better than the
formulation of Newton because it explains more accurately or more simply our
3 4 5 6 7 observations of an assumed, external, immutable N a t u r e — a "Nature behind the
Knob Setting veil." To emphasize the difference in the engineer's and the scientist's notions
Figure 4 Objective Function vs. Setting of best, consider the odd situation that would arise if time and space turned out
to be h u m a n - m a d e myths. I grant that for most of us this is an unbelievable sit-
uation, b u t some philosophers believe just that. For example, in the book The
bined g r a p h is given in Figure 4. The bottom two curves are the return func- Tao of Physics by Fritjof Capra, Swami Vivekanada is quoted as saying, "Time,
tions for the conflicting criteria that w e r e s h o w n in Figure 2 and Figure 3. The space, and causation are the glass through which the Absolute is seen. . . . In
u p p e r d o t t e d c u r v e is the s u m of these two u n d e r the assumption that im- the Absolute there is neither time, space or causation." With no external time
p r o v e m e n t s in both picture and s o u n d are equally desirable. Verify that this and space to refer to, the scientist's best, as an approximation to these concepts,
c o m b i n e d d a s h e d curve w i t h a value of 12 at setting 4 is equal to our previous becomes vacuous. On the other hand, the engineer's best is still a valid model
reading of 8 a n d 4 at the same point. This d a s h e d curve is the c o m m o n m e a s u r e of society's (now) mythical concept of time and space. Best for the scientist im-
of g o o d n e s s for the problem. It is s o m e t i m e s called the objective function or mea- plies congruence with an assumed external Nature; best for the engineer implies
sure of system effectiveness. We w a n t the largest or m a x i m u m value of the objec- congruence with a specific view of Nature.
tive function. This n u m b e r is called the optimum or best value. For our television
The a p p r o p r i a t e view of Nature for optimization by the engineer is not just
set the o p t i m u m setting c o r r e s p o n d s to a control setting of the knob or m a n i p -
an objective, faithful model of society's view, but includes criteria known only
ulated variable of 4. to him. O n e important consideration in lowering the cost of an automobile, for
O p t i m i z a t i o n is a c o m p r o m i s i n g process. If the setting is below 4, the sound instance, is its ease of manufacture. If standard parts can be used or if the auto-
is better a n d the picture worse, b u t in such a w a y that our net satisfaction with mobile can be constructed on an assembly line instead of by hand, the cost of
the c o m b i n a t i o n is lower. Likewise, a setting above 4 p r o d u c e s a better picture each unit goes d o w n . Ease of manufacture is a criterion seldom considered by the
but this is m o r e than c o m p e n s a t e d for by a reduction in the fidelity of the s o u n d , public w h e n it decides which automobile to b u y b u t one that is essential to an
and again o u r satisfaction is less. From an engineering point of view point 4 is accurate model, for determining best automobile design. To quote a prominent
the best, or optimal, setting. The engineer calls the process of balancing an im- engineer Karl Willenbrock from an editorial in Science magazine, "Their [engi-
p r o v e m e n t in one criterion against a w o r s e n i n g of another a trade-off. O n e can- neers'] designs must satisfy scientific as well as nonscientific criteria such as
not h a v e the best of all possible w o r l d s — i n this case the world of sight and the manufacturability, maintainability, risk-minimization, a n d cost-effectiveness."
world of hearing. The most one can hope for is the best in the combined, real Because of these additional variables, the appropriate optimization model is not
world. just a surrogate for society but includes subjective considerations of the engi-
A l t h o u g h the preceding e x a m p l e is simple, it contains the important theo- neer w h o makes the design.
retical aspects of optimization theory. If the engineer can develop a m a t h e m a t - In general, the o p t i m u m is different w h e n an optimization space with a re-
ical m o d e l for his problem including the a p p r o p r i a t e optimization space, return duced n u m b e r of criteria is used. The best automobile from the points of view
functions, a n d c o m m e n s u r a t i o n , the o p t i m u m is determined. This does not mean of the public and the engineer will therefore differ. The person w h o criticizes
that it is easy to find, but only that in principle it can be found. The design of the engineer for not providing an automobile to last fifty years makes the error
an a u t o m o b i l e will have an optimization space with a m u c h larger n u m b e r of of not using a complete axis system. She is almost certainly not considering the
conflicting criteria such as fuel economy, p o w e r , safety, comfort, and style, and ease with which such an automobile could be manufactured, and so forth. Her
it will h a v e m o r e complicated return functions. As a result, a large c o m p u t e r view of the true o p t i m u m in this reduced axis system is as deficient as the knowl-
a n d sophisticated numerical strategies w o u l d be n e e d e d to obtain the o p t i m u m edge of a three-dimensional world to the two-dimensional inhabitants confined
if w e w a n t e d to calculate it analytically. But the important point is that given a to live in a two-dimensional world in Abbot's classic novel Flatland. The design
valid m o d e l , evaluation of best is a well-defined engineering operation. of a long-life automobile is possible a n d would be as exciting a challenge to the
Unlike science, engineering d o e s not seek to model an a s s u m e d , external, im- engineer as the present line of products. But the d e m a n d for one in the United
m u t a b l e reality, b u t society's perception of reality including its myths and prej-
20 / DISCUSSION OF THE METHOD Some Thoughts on Engineering / 21

States is so low, and the cost of p r o d u c i n g it so high, that the cost per vehicle
w o u l d be prohibitive. By the w a y , such a car, the Rolls-Royce, is actually on to-
d a y ' s m a r k e t , b u t if all automobiles in America cost as m u c h as a Rolls-Royce,
few A m e r i c a n s w o u l d n e e d a two-car garage.
As a second e x a m p l e of a deficient system of axes being used by s o m e m e m -
bers of the public, consider the complaint communication engineers occasion-
ally hear, "This holiday season all of the telephone lines were busy and I couldn't
get t h r o u g h . You w o u l d think the people at the telephone c o m p a n y could an-
ticipate the rush." Again w e h a v e two different axis systems being u s e d — o n e
by the layperson a n d a different one by the communication engineer. The engi-
neer could easily design a telephone system for the busiest period of the year,
but the extra e q u i p m e n t that w o u l d be needed w o u l d remain idle the rest of the
4 5 6 7
year a n d w o u l d h a v e to be stored, insured, a n d maintained. The engineer uses Knob Setting
an axis along w h i c h the cost of the extra, seldom-used e q u i p m e n t is being traded Figure 5 Revised Weighting
off against the loss of service. I agree that the public has a right (I w o u l d say ob-
ligation) to help select the p r o b l e m s for solution, the major criteria to be included
in the design, a n d the relative importance of these criteria. The engineer, h o w -
struction is begun. From a practical point of view neither can be determined per-
ever, a l w a y s uses additional criteria u n k n o w n to the public. Since in general the
fectly. Assume, however, that s o m e h o w we know these quantities. To use our
o p t i m u m will shift d e p e n d i n g on the n u m b e r of variables used, it is naive for
theoretical optimization procedure, the separate optimization spaces for all of
the layperson to criticize the engineer's final design without justifying w h y the the individuals that make up society m u s t now be combined to produce a rep-
reduced set w a s more a p p r o p r i a t e . resentative average. No unique, completely defensible average exists. It will de-
Theoretically, then, best for an engineer is the result of m a n i p u l a t i n g a m o d e l p e n d on whether a numerical, political, or economic scheme is used to weight
of society's perceived reality including additional subjective considerations the individual optimization spaces to p r o d u c e it. In practice the engineer must
k n o w n only to the engineer constructing the model. In essence, the engineer cre- determine this average. Assume, however, that this has also been done. At last
ates w h a t he thinks an informed society should w a n t based on his k n o w l e d g e we are able to evaluate the theoretical best for society, b u t unfortunately we now
of w h a t an u n i n f o r m e d society thinks it w a n t s . have no assurance that this average is acceptable to any one member of the
A l t h o u g h the theoretical p r o c e d u r e described here is useful for indicating g r o u p . If one-half of the public wanted a vehicle with four wheels and the other
w h a t a n engineer m e a n s by best, it r u n s into serious practical difficulties. Accu- half preferred one with two, I doubt the engineer should design a three-wheeled
rate k n o w l e d g e of the optimization space a p p r o p r i a t e for society is often un- car. Or as one of my former chemical engineering professors* p u t it memorably,
available, and, even if it is available, it does not necessarily reflect the desire of although a bit earthily, "Always remember that the engineer designs for the in-
any o n e individual. Society is an abstraction for a g r o u p of individuals each of dividual, not the average. The average American w o u l d have one tit and one
w h o m h a s her o w n personal definition of best. This abstraction is necessary be- ball."
cause a specific television set is to be p r o d u c e d , which implies one optimization
Although the theoretical procedure for determining the best design is well
space, o n e set of r e t u r n functions, a n d one set of weighting coefficients. If w e
defined, the practical p r o c e d u r e is not. In some completed engineering projects
were to ask different people to adjust the television set that w a s used in the ear-
we have experimental evidence that the axis system ultimately chosen as rep-
lier e x a m p l e each w o u l d choose a different setting of the knob. A person w h o
resentative of society was deficient. The San Francisco Embarcadero, originally
w a s h a r d of hearing or interested in high fidelity w o u l d place more e m p h a s i s
begun in 1959, has become a classic example of the practical problem of trying
on s o u n d than picture, a n d her o p t i m u m setting w o u l d shift.
to evaluate society's o p t i m u m . It was designed as the best way to move traffic
Figure 5 is d r a w n with the quality of the picture half as important as that of
about the city. Money was appropriated and construction was begun. But the
the s o u n d . For convenience the return function for the sharpness of the picture
Embarcadero is now k n o w n as the "freeway that goes n o w h e r e . "
has been r e d r a w n r e d u c e d by one-half so that the combined curve can be cal-
The reason is shown in Figure 6, w h e r e we see the end of an uncompleted
culated w i t h ease. Verify that the value of the objective function at a setting of
ramp. It w a s abandoned in mid-construction because the design failed to in-
4 is n o w equal to 8, or one-half of the old value for sharpness (4) plus the old
clude considerations that ultimately proved important. Criteria such as "Don't
value for fidelity (also 4). As expected the o p t i m u m has n o w shifted to a lower
increase the noise level or density of people in my n e i g h b o r h o o d " and "Don't
setting—one that favors s o u n d . Each person will choose a different value d e -
decrease the overall quality of life" were important to the citizens of San Fran-
p e n d i n g u p o n the relative importance of s o u n d and picture to her. In this ex-
cisco. Too expensive to tear d o w n , the Embarcadero stood for many years as a
a m p l e it w o u l d be easy to d e t e r m i n e the best setting for each individual exper-
m o n u m e n t to the difference between engineering theory and the real world. To
imentally b v simply asking her to set the dial as she desires. But in actual practice
quote an article in the Los Angeles Times (November 5, 1985):
the engineer m u s t d e t e r m i n e w h a t best m e a n s for each individual before the
p r o d u c t is even built. Therefore, he must guess what criteria each individual
considers important a n d the relative importance attached to each before con- 'Prof. Van Winkle.
 Some Thoughts on Engineering / 23
22 / DISCUSSION OF THE METHOD
Study for Suite 3 (4 Desks) Two Sets of Bunk Beds

Bathroom for - Living Room for

uites 1-4

Top View of Typical Floor

Figure 7 Mystery Floor Plai

corners. In addition, the w i n d o w s on the extreme outside were reduced to tiny

Figure 6 San Francisco Embarcadero
slits similar to those used in castles to protect archers defending the fortress.
This innovation was evidently m a d e to minimize the air conditioning cost by
reducing the w i n d o w area. The overall floor plan looks strikingly like a beehive,
Politicians here have for 20 years won hearts—and voles—bv pledging to de-
and well it might. The beehive has evolved over centuries to have m a x i m u m ca-
molish an ugly, unfinished stump of the elevated, double-deck Embarcadero
pacity (for honey) for a m i n i m u m a m o u n t of construction materials (since bees
Freeway along the city's waterfront. On Monday, the Board of Supervisors
have to transport the construction materials in m a n y trips.)
voted 8 to 2 to proceed with preliminary engineering on a project that would
The history of these buildings is revealing and serves as experimental evi-
exchange the freeway for a tree-lined boulevard and better mass transit. . . .
dence that engineers d o not always get the optimization space right. Originally
For years, the freeways—especially the Embarcadero—were seen as useless
these skyscrapers were to serve as m o d e r n , desirable, state-of-the-art dormito-
eyesores that went nowhere and blocked views.
ries, but w h e n students refused to live there by choice, they had to be made into
required freshmen d o r m s . I have been told that when even this did not produce
Evidently, the $10 million axis system left out a criterion that w a s important to
a satisfactory occupancy rate, with poetic justice, at least one of them was m a d e
the voters. into an administrative building. These buildings were simply not optimal as
As an additional example, r e m e m b e r the u r b a n renewal projects in N e w York demonstrated bv the exodus of students. Although theoretically the best design
City that w e r e designed to be optimal with respect to m o n e y , space utilization, is determined once the optimization space is known, practically it is hard to be
and materials. They p r o v e d to be suboptimal w h e n the tenants broke w i n d o w s , sure that we have not neglected an important axis in constructing this space. In
d u m p e d garbage in the halls, a n d m o v e d out. all of these three examples, the optimization space of the engineer proved in
Figure 7 serves to introduce a third and final example of an engineering proj- practice to be a poor representation of the one preferred by society.
ect that t u r n e d out to be s u b o p t i m a l . It depicts the schematic of the floor plan Please remember the engineer's notion of best or o p t i m u m well. By the time
used in t w o skyscrapers built on the c a m p u s of a w e l l - k n o w n (but charitably we have completed Part IV it should become obvious that it is preferable to the
u n n a m e d ) university in the N o r t h . Beginning at the center of the figure, w e note classical Platonic one in defining the universal method. The best zee can do is not
the elevator a n d utility areas s u r r o u n d e d by six living areas. Each living area the best; the best we can do is the best we can do.
consists of a b a t h r o o m a n d living room for four separate suites. Each suite has A fundamental characteristic of an engineering solution is that it is the best
a s t u d y area with four desks for the four inhabitants w h o sleep in b u n k b e d s in available from the point of view of a specific engineer. If this engineer knew the
the outer area. In all, sixteen people are h o u s e d in each of the six living areas. absolute good, he would do that good. Failing that, he calculates his best based
It is also w o r t h noting that n o n e of the walls meet in right angles. Psychologists on his subjective estimate of an informed society's perception of the good. With
' "' ' -" u"<-'^'-"" <-HunrH>nrprl w h e n thev are in rooms with odd
24 / DISCUSSION OF THE METHOD Some Thoughts on Engineering / 25

d o u b t about the criteria that are important to society, with d o u b t about the rel- President Kennedy's exhilarating problem statement before Congress was the
ative i m p o r t a n c e of these criteria, a n d with d o u b t as to w h e t h e r society's best g u n that sounded the start of one of America's most spectacular engineer-
reflects the individual's best, h o w can the engineer design the o p t i m u m prod- ing races. Make no mistake: It w a s primarily a race between engineers, not
uct? W h a t strategy does he use? scientists.
Similarly, when Congress passed a n d President Johnson signed the Ap-
Uncertainty palachian Regional Development Act, an engineering problem was implied:
Not m u c h n e e d s to be said about the last characteristic of an engineering prob-
This bill authorized a total of $840 million in federal grants over the six-year
lem, uncertainty, because we h a v e seen its specter p e r v a d e each of the previous
period ending June 30, 1971, to pay up to 70 percent of the cost of building
three characteristics. Indeed this uncertainty, this d o u b t , this struggle to solve a 2,350 miles of development highways and 1,000 miles of local access roads in
p r o b l e m in the absence of complete information is one of the identifying signa- the region.
tures of an engineering problem. Until s o m e o n e can r e m o v e this uncertainty, I,
along w i t h m y engineering colleagues, m u s t continue our struggle to cause the
Reagan, Kennedy, Johnson—in these three examples a president asked for the
best c h a n g e in an uncertain situation within the available resources. W h a t strat-
best change from an initial state A to a different, more desirable state B for a
egy s h o u l d w e use?
complex, poorly understood system within the resources at his disposal. For
each, he needed the services of people w h o u n d e r s t o o d the engineering method.
Engineering change is not limited to the creation of physical devices such as
EXAMPLE ENGINEERING PROBLEMS defense systems, spaceships, or h i g h w a y s . Consider the following statement:

Best, change, uncertainty, and resources—although w e d o not as yet know w h a t We . . . have already declared war on poverty in all its forms. . . . We have
the engineer's strategy for causing change is, it should not be too difficult to rec- $8,700,000 going into vocational training of 3,000 persons for 50 occupations.
ognize a situation calling for its use. But unfortunately it is. W h e n President Rea- We have $4 million in loans and grants for housing. We have surplus food dis-
gan p r o m o t e d a new generation of space w e a p o n s to create a defensive umbrella tribution in all the eastern Kentucky counties. We have 360 schools, 10,000
for America a n d then called on the "scientific c o m m u n i t y " to give us a w a y of pupils, participating in the school lunch program. Under our community work
d e v e l o p i n g it, he confused science and engineering. Relatively speaking, little and training program, we have 48 programs going, 1,400 men in action.
n e w science is involved. N e w t o n ' s law of gravitation, the equations of motion,
a n d the theory of energy emission by lasers or particle b e a m s are all reasonably Once again, we sense the concepts best, change, uncertainty, and resources all used
well u n d e r s t o o d by the scientist. If such a device is to be developed, the presi- as an engineer would use them; once again, President Johnson should have called
d e n t w o u l d have been better advised to call on the "engineering c o m m u n i t y . " on the engineering c o m m u n i t y for help.
Journalists share this same confusion about w h a t constitutes a scientific prob- Other political, economic, and psychological examples that require the tal-
lem a n d w h a t constitutes an engineering one. W h e n reporters seeking informa- ents of the engineer are easy to find. Some are almost trivial, others are more
tion about Reagan's project w e n t to "scientific experts" to evaluate the "feasi- complex—but all have the characteristics of an engineering problem. Perhaps a
bility of this space-age missile defense system," they w e n t to the w r o n g place. politician w a n t s to be reelected or to w i n s u p p o r t in Congress for the construc-
Its feasibility is certainly m o r e in d o u b t because materials able to survive the tion of a d a m in her home district; p e r h a p s an economist w o u l d like to increase
tensile stresses, radiation d a m a g e , a n d alien t e m p e r a t u r e e n v i r o n m e n t or be- the gross national product or find a way to reduce the national debt; perhaps a
cause a device able to maintain the severe pointing accuracy needed cannot be psychologist would like to stop a child from biting her nails or condition a race
found than it is because s o m e t h i n g that violates the k n o w n laws of N a t u r e will to create a Utopian state using behavioral engineering. The changes implied by
be found. If feasibility is the question, journalists should contact the local d e a n these examples are usually not associated with the engineer, but careful study
of a college of engineering, not their resident scientist. of the characteristics they share with the obvious engineering projects of de-
Since confusion evidently exists in the m i n d of the nonengineer as to w h a t signing a nylon plant, constructing a bridge across the Mississippi River, and
constitutes an engineering p r o b l e m , let us consider several additional examples building an electrical p o w e r station for N e w York City shows a definite pattern.
w i t h the defining characteristics of one in mind. For each, the engineering m e t h o d is needed.
The statement of an engineering p r o b l e m might well be: If you, as with all h u m a n s since the birth of h u m a n k i n d , desire change; if the
system you want to change is complex and poorly understood; if the change
I believe that this nation should commit itself to achieving the goal, before this you will accept must be the best available as you balance often conflicting cri-
decade is out, of landing a man on the moon and returning him safely to the teria; and if it is constrained by limited resources, then you are in the presence
earth. No single space project in this period will be more impressive to of an engineering problem. If you cause this change using the strategy described
mankind, or more important for the long-range exploration of space; and none next, then you are an engineer.
so difficult or expensive to accomplish. . . . [The cost would be] $531 million
in 1962 and an estimated $7-9 billion over the next five years.