The theory of decisions under uncertainty share basic assumptions with system dynamics. Both methods require that decisions are based on only available information, and both methods focus on the development of policy rules that improve system performance. Both methods have other implications for parameter estimation than conventional deterministic analysis. Fluctuations are frequently studied in system dynamics, and fluctuations and randomness are of great importance for decisions under uncertainty. Decisions under uncertainty can be studied by analytical methods, dynamic programming and Monte Carlo simulations. The latter method is quite easily applied to system dynamics models. Using Monte Carlo simulations we show that uncertainty has important implications for decisions influencing the “greenhouse” effect. Note that risk aversion is not an issue in this example. The theory of decisions under uncertainty brings new qualitative insights to system dynamics, and facilitates quantitative improvements of policy rules. Referring to or applying the theory of decisions under uncertainty might help to get a wider academic acceptance of system dynamics models, which are often thought of as being realistic but quite uncertain. The principles of system dynamics might bring the field of decisions under uncertainty in the direction of greater realism. The focus on real life interpretation of system dynamics models is most useful for the application of apriori information. Apriori information is needed to establish important autocorrelation in cases where short time-series do not contain sufficient information.