Bayesian forecasting for complex systems using computer simulators

Although computer models are often used for forecasting future outcomes of complex systems, the uncertainties in such forecasts are not usually treate d formally. We describe a general Bayesian approach for using a computer mo del or simulator of a complex system to forecast system outcomes. The appro ach is based on constructing beliefs derived from a combination of expert j udgments and experiments on the computer model. These beliefs, which are sy stematically updated as we make runs of the computer model, are used for ei ther Bayesian or Bayes linear forecasting for the system. Issues of design and diagnostics are described in the context of forecasting. The methodolog y is applied to forecasting for an active hydrocarbon reservoir.