Many of today's pressing policy challenges—such as climate change and inequality—are characterized as wicked problems. How might decision making under deep uncertainty be used to demonstrate methods that may help resolve the tension between differing approaches for addressing these problems?
Robust Decision Making
Achieving Tomorrow’s Goals Across an Uncertain Future
How might different short-term actions affect long-term outcomes? With sufficient knowledge and having unknowns well-characterized by probabilities, such decisions over choices are amenable to the tools of analysis. But when information is sparse or unavailable and probability estimates unreliable, these tools may be inoperable or their results misleading. We are then left to weigh alternative stories: “What if..? Suppose that..? Could this..?”. Instead of the rich tool kit of analytical methods for deductive reasoning, what remains are competing, unsystematic narratives.
The individual ability to reason over “What if...?” stories becomes even more challenging when a group (or assembly of groups) tries to reason collectively through complex and uncertain futures. Business and government processes become overburdened when confronted by an environment not imagined by their framers.
Robust Decision Making (RDM) is a method designed to supply the missing machinery for systematic, shareable reasoning and decision making under conditions of deep uncertainty (DMDU). RDM’s rigor comes from using the same models already being used to conduct more traditional analysis. The difference lies in the use of those models. Rather than seeking to enhance the ability to be predictive – unlikely to prove successful under deep uncertainty – RDM supports the systematic construction, testing and selection of short-term actions that will be consistent with long-term goals over many alternative futures. (That is, rather than decisions optimized for planning assumptions that may not anticipate how the future actually unfolds, robust decisions will achieve set threshold for indicators of satisfactory outcomes across a wide range of plausible futures.)
Robust Decision Making (RDM) is a widely used approach for Decisionmaking Under Deep Uncertainty (DMDU). Originally developed at the RAND Corporation, RDM asks “How can we make good decisions without first needing to make predictions?”
RDM focuses stakeholders’ attention on the characteristics of their policy options rather than on predictions of the future. Through an iterative process, stakeholder deliberation informs the kinds of analysis that are needed to answer key questions about the policy problem, and the analysis provides information over which stakeholders deliberate.
Developing RDM Methods
RAND has played a leading role in the development of Robust Decision Making, beginning with early research on exploratory modeling, to the first report-length description and application of Robust Decision Making, and subsequent academic journal articles formalizing the method, explaining how it can be applied, and describing key techniques. Below are more recent RAND publications on RDM methods.
Applying RDM to Policy and Decisionmaking
The RAND Center for Decision Making under Uncertainty not only works to develop RDM as a methodology, we also apply it to a wide range of disciplines and problems.