The systematic problem solving thread illustrates the diverse views held
regarding problems and the processes for solving them. I'd like to add to
and clarify/modify some of what has been posted so far.
My own view, shaped by the writings of folks like John Dewey, Charles Kepner
and Benjamin Tregoe, and Messrs. Shaw, Newell and Simon, is that a problem
exists whenever a requirement for action is coupled with uncertainty
regarding the action to take. As I once read, "Problem solving is what you
do when you don't know what to do." Sometimes the requirement for action is
rooted in a clearly defined gap between "what is" and "what should be." On
other occasions, the current situation is simply untenable and part of the
difficulty lies in identifying some new goal state. In all cases, however,
there is uncertainty regarding action.
Problem solving, or figuring out what to do, serves functionally to reduce
uncertainty. It is, therefore, an information-based search activity. The
search for information can be systematic, haphazard, creative, flexible, or
rigid. (Insert your own adjectives here.) Uncertainty, of course, attaches
to people, not situations. The uncertainty to be reduced is that of the
person(s) with the responsibility for the situation in which action is to be
taken so as to alter that situation. The reduction of uncertainty is in fact
a speculative undertaking, that is, we come up with a course of action we
think or believe will work. There remains the sometimes nasty business of
actually doing it--in a word: implementation.
Depending on the scope, scale and complexity of the course of action being
undertaken, one, a few, dozens, or even hundreds of people must understand
the course of action. Some so as to have their uncertainty reduced, too, and
some simply so they can "go along with the program." Hence, the need to have
buy-in through involvement on the part of some and the need to "sell" to
others.
People in organizations do all these things--and more--and they do them
fairly well. If they were lousy at it, organizations would collapse with
alarming frequency. People are pretty good problem solvers. But they vary
widely in their abilities, skills, preferences, mastery of techniques, and so
on. Some rely heavily on creative approaches, while others favor a more
systematic approach. Some know to engage the implementers early on and some
don't. Some know that engaging the implementers is a generally useful thing
to do but not always necessary--and some don't. Some believe in "root"
causes and set out to identify them; some don't believe in root causes--or in
any other instance of cause-and-effect reasoning. On and on the variation
goes.
Yet, in the last analysis, solving problems is about changing things, it is
about intervening. To intervene is to change things purposefully, that is,
with some goal or outcome in mind. Change in organizations is almost always
indirect; that is, you don't change it (e.g., profit), you change something
else (e.g., cost) and it changes as a result. Consequently, if we are to
intervene successfully, we must be able to say that a given action produces a
given result and, conversely, that a given result can be produced by some
range of given actions. Being able to say these things with confidence and
having our confident statements subsequently borne out by experience, entails
a knowledge of the structure of the situation in which we are intervening.
In other words, we must be able to show how intervening over here will
produce the desired result over there or, conversely, how a given result over
there can be produced by actions over here.
Typically, we represent the structure of the situations in which we are
intervening by way of diagrams (e.g., flowcharts, treecharts, and other
"structural" drawings). Generally speaking, I find three basic classes of
diagrams to cover the vast majority of problems encountered in the work
place. One for people, one for processes, and one for profits. Said a
little differently, these relate to problems of human behavior and
performance, functional and organizational performance, and financial as well
as other measurement-based results. Clearly, the more interesting problems
found in organizations involve all three classes of problems, intertwined in
what are known as "messes" (i.e., sets of related but as yet undifferentiated
problems).
The one thing I don't believe in is the so-called "wicked problem." All
problems can be tamed.
Regards,
Fred Nickols
Executive Director
Strategic Planning & Management Services
Educational Testing Service
fnickols@ets.org