Robust Decision Making: Coping with Uncertainty

Document Sample
Robust Decision Making: Coping with Uncertainty Powered By Docstoc
					a good model is established, market-      vides a great deal of clarity, since it          ten for earlier versions of the Millennium
ers can identify which attributes to      requires answers to the question of              Project’s Futures Research Methodology
stress or improve in order to increase    what’s important. It also promotes               CD-ROM by several contributors at The
                                                                                           Futures Group, a multinational strategy con-
market share.                             thinking about what can go wrong.
                                                                                           sulting firm founded by Theodore J. Gordon
   For complex decisions that may af-                                                      (now senior fellow of the Millennium Project).
fect many people for long periods of      About the Authors                                The Futures Group International, can be
time, the simple utility matrix pro-      The original version of this article was writ-   reached at www.futuresgroup.com.




By Robert J. Lempert, Steven W. Popper, and Steven C. Bankes

Robust Decision Making: Coping with Uncertainty
Predicting the future and                 the question of how one can use im-              futures ought to be considered. It can
                                          perfect computer models to inform                help decision makers avoid “over-
then acting on our predic-                policy decisions, particularly to deal           arguing,” which occurs when deci-
                                          with the next wars rather than previ-            sion makers pretend they are more
tions leaves us vulnerable to             ous ones.                                        certain than they actually are to
                                             In brief, RDM uses the computer               avoid losing credibility in policy de-
surprises. So we need deci-               to support an iterative process in               bates — by allowing them to ac-
                                          which humans propose strategies as               knowledge multiple plausible
sions that will work in a va-             potentially robust across a wide                 futures and to make strong argu-
                                          range of futures. The computer then              ments about the best policies for
riety of potential situations.            challenges these strategies (stress              hedging against a wide range of con-
                                          tests) using simulations and data ex-            tingencies.
   Robust decision making (RDM) is        trapolations to suggest futures where               Computer-supported RDM at its
a framework for making decisions          these strategies may perform poorly.             root combines the best capabilities of
with a large number of highly imper-      The alternatives can then be revised             humans and machines. Humans
fect forecasts of the future. Rather      to hedge against these stressing                 have unparalleled ability to recog-
than relying on improved point fore-      futures, and the process is repeated             nize potential patterns, draw infer-
casts or probabilistic predictions,       for the new strategies.                          ences, formulate new hypotheses,
RDM embraces many plausible                  Rather than first predicting the              and intuit potential solutions to
futures, then helps analysts and deci-    future in order to act upon it, deci-            seemingly intractable problems. Hu-
sion makers identify near-term ac-        sion makers may now gain a system-               mans also possess various sources of
tions that are robust across a very       atic understanding of their best near-           knowledge — tacit, qualitative, expe-
wide range of futures — that is, ac-      term options for shaping a long-term             riential, and pragmatic — that are not
tions that promise to do a reasonable     future while fully considering many              easily represented in traditional
job of achieving the decision makers’     plausible futures. The result is near-           quantitative formalisms. Working
goals compared to the alternative op-     term policy options that are robust              without computers, humans can of-
tions, no matter what future comes        — i.e., that, compared to the alterna-           ten successfully reason their way
to pass. Rather than asking what the      tives choices, perform reasonably well           through problems of deep uncer-
future will bring, this methodology       
				
DOCUMENT INFO
Description: Robust decision making (RDM) is a framework for making decisions with a large number of highly imperfect forecasts of the future. Rather than relying on improved point forecasts or probabilistic predictions, RDM embraces many plausible futures, then helps analysts and decision makers identify near-term actions that are robust across a very wide range of futures -- that is, actions that promise to do a reasonable job of achieving the decision makers' goals compared to the alternative options, no matter what future comes to pass. The strength of robust decision making is its flexibility. The four basic steps in robust decision making are as follows: 1. Consider ensembles of large numbers of scenarios. 2. Seek robust, rather than optimal, strategies that do "well enough" across a broad range of plausible futures. 3. Employ adaptive strategies to achieve robustness. 4. Use computer tools designed for interactively exploring multiple plausible futures.
BUY THIS DOCUMENT NOW PRICE: $6.95 100% MONEY BACK GUARANTEED
PARTNER ProQuest LLC
ProQuest creates specialized information resources and technologies that propel successful research, discovery, and lifelong learning.