Computer Assisted Reasoning for Robust Strategies

Decision makers often confront deep uncertainty where standard decision support tools fail to provide support. Sophisticated planners seek strategies robust across a wide range of plausible scenarios. But most analytical tools attempt to optimize for aparticular future lending little aid in the search for robust minimal regret strategies. Robustness is more challenging computationally than optimization. This has left a large gap between intuition and analytical support. However, recent advances incomputer technology have made possible a revolutionary new generation of decision support tools. In this project, Evolving Logic will use its experience with its Computer Assisted Reasoning r technology to build a prototype system for robust, adaptiveplanning under conditions of deep uncertainty. The prototype will be generalizable to specific applications combining the best features of scenario planning with traditional quantitative approaches. Means for rapidly adapting this technology to a rangeof military and commercial applications will also be developed. The result will be an unprecedentedly powerful yet intuitive way for decision-makers, even those with little experience or quantitative facility, to design and understand the strengths andweaknesses of various plans in full view of the inherent uncertainties and for groups of planners to draw inferences and gain insights from within existing planning and decision-making institutions.