Robust Decision Making and Scenario Discovery in the Absence of Formal Models

Robust decisionmaking (RDM) is a method for aiding decision making under deep uncertainty that uses models not as predictors but as generators of cases exploring assumptions and outcomes. RDM was intended for use with formal models. However, we show a model-less RDM application to a portfolio planning problem (selecting U.S. Army security cooperation activities with a partner country) seeking to achieve several objectives. In the absence of formal models, the analysis tests candidate actions against different explicit statements of causal relationships to allow more systematic reasoning over choices and outcomes. Doing so renders assumptions about complex systems explicit, characterizes uncertainties in terms of effect on weighting policy choices rather than as presently unknowable probabilities, provides a venue for planners and evaluators to share findings and insights, and yields explicit expressions of theories of causation (TOC) that themselves act as formal models where none had previously existed. RDM allows important scenarios for decision planning to be generated analytically.This may address for several scenario applications the question of determining important futures and also how scenario results may be used directly to inform policy and operational decisions.