Robust Decision Making

Shaping the Future

The 2003 book Shaping the Next One Hundred Years: New Methods for Quantitative, Long-Term Policy Analysis lays out four key elements for successful long-term policy analysis:

  • Consider large ensembles of plausible futures, not a single best estimate forecast
  • Seek robust strategies that perform well across many plausible futures, not a strategy optimal for one particular view
  • Employ adaptive strategies, ones that evolve over time in response to changing conditions
  • Use the computer as a “prosthesis for the imagination,” not a calculator

Chapter 3 of this book describes robust decision-making (RDM), a quantitative policy analytic method that captures these key elements.

Many Pardee Center projects use this RDM approach.

Shaping the Future — March 2005

Scientific American

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

This article provides a brief summary of RDM.

Read More at sciam.com

Thinking inside the box: A participatory, computer-assisted approach to scenario discovery — Oct. 2009

computer design, courtesy of Flickr

Scenarios provide a commonly used and intuitively appealing means to communicate and characterize uncertainty in many decision support applications, but can fall short of their potential especially when used in broad public debates among participants with diverse interests and values. This paper in the journal Technology Forecasting and Social Change describes a new approach to participatory, computer-assisted scenario development that we call scenario discovery, which aims to address these challenges. This paper systematically describes the scenario discovery concept and its implementation, presents statistical tests to evaluate the resulting scenarios, and demonstrates the approach on an example policy problem involving the efficacy of a proposed U.S. renewable energy standard. The paper also describes how scenario discovery appears to address several outstanding challenges faced when applying traditional scenario approaches in contentious public debates.

Read More

Shaping Tomorrow Today: Near-Term Steps Towards Long-Term Goals — Aug. 2009

Pardee Workshop, Jan. 2009

In March 2009, the RAND Pardee Center hosted a workshop called “Shaping Tomorrow Today: Near-Term Steps Towards Long-Term Goals.” The workshop gave policymakers and analysts an opportunity to explore new methods and tools that can help improve long-term decisionmaking. The proceedings from this important event are now available.

Full Document

Managing the Risk of Uncertain Threshold Responses: Comparison of Robust, Optimum, and Precautionary Approaches — Oct 2007

Risk Analysis

Robert J. Lempert and Myles T. Collins

This study uses a simple computer simulation model to compare several alternative frameworks for decision making under uncertainty–optimal expected utility, the precautionary principle, and three different approaches to robust decision making–for addressing the challenge of adding pollution to a lake without triggering unwanted and potentially irreversible eutrophication.

Read More at interscience.wiley.com

Scenario Discovery Toolkit Now Available on the Comprehensive R Archive Network (CRAN) — Mar. 10, 2008

image courtesy of Flickr/ Wonderlane

Scenario discovery supports robust decision making by identifying scenarios that represent the vulnerabilities of proposed policies.

Learn more about scenario discovery Toolkit on public domain (non-RAND) CRAN site

We welcome your input. Send suggestions for links and other ideas to lempert@rand.org.

My RAND ?

Saved Items

Recommended

View My Profile