When faced with a significant risk, society must decide howmuch to invest in prediction and response. For example, in the face of hurricane risk how much should we invest in better forecasting versus increased evacuation speed? To address this need, we develop a Markov decision processes model to analyze the interaction between the emergency response system and the emergency forecasting system. The model shows the tradeoff between the two investments given a budget limit. In addition, the research indicates that the superiority of the investment changes sharply by the lead time.
Available at: http://works.bepress.com/eric_bickel/6/