From Prediction to Implementation: Results Chains with Bayesian Networks for Decisions Under Uncertainty
Abstract
Decision modeling normally consists of estimating the expected utility of alternatives under consideration. The alternative with the best overall performance in terms of expected utilities of the outcomes is recommended for implementation. Final decisions are often made taking this guidance into account−with additional considerations outside the predictions made by the decision model. After decision selection, the implementation phase of decision making begins following additional tools and frameworks, such as adaptive management. Powerful but underutilized implementation-aiding tool for complex decisions are results chains. Results chains were first pioneered by the Foundations of Success in the 1990s (Foundations of Success. 2009. Using results chains to improve strategy effectiveness: an FOS how-to guide. Foundations of Success, Bethesda, Maryland, USA) and further described for environmental management by Margoluis and colleagues (Margoluis, R., et al., 2013. Results chains: a tool for conservation action design, management, and evaluation. Ecology and Society, 18(3).) Results chains provide conceptual models connecting decisions to intermediate results to ultimate outcomes for the decision. Even when qualitative, results chains provide insights into implementation when there are large uncertainties about the outcomes of decisions. This presentation will further describe results chains with several past conservation case studies of their usage. A process for developing quantitative results chains based on Bayesian network intervention models will be described. Coupling results chains with Bayesian decision networks may provide effective support for achieving desirable outcomes from decisions made under uncertainty.
About the Presenter
John Carriger is an independent research scientist and consultant. John has a marine science Ph.D. from the College of William and Mary. John’s research interests include applying risk assessment, decision analysis, and weight of evidence tools to environmental problems.