This paper includes a very handy review of theoretical tools that help with dynamic decision-making:

Dynamic decisions arise in many applications including military, medical, management, sports, and emergency situations. During the past 50 years, a variety of general and powerful tools have emerged for understanding, analyzing, and aiding humans faced with these decisions. These tools include expected and multi-attribute utility analyses, game theory, Bayesian inference and Bayes nets, decision trees and influence diagrams, stochastic optimal control theory, partially observable Markov decision processes, neural networks and reinforcement learning models, Markov logics, and rule-based cognitive architectures. What are all of these tools, how are they related, when are they most useful, and do these tools match the way humans make decisions? We address all of these questions within a broad overview that is written for an interdisciplinary audience. Each description of a tool introduces the principles upon which it is based, and also reviews empirical research designed to test whether humans actually use these principles to make decisions. We conclude with suggestions for future directions in research.