In this paper, a new methodology based on system modelling and optimization is developed to determine forest fire risk, and subsequent resource allocation, over a regional area. A forest fuel moisture model and a propagation model are developed, in order to determine the dynamic risk assessment. The objective of resource allocation is twofold. In the preventive phase means and crews are re-allocate on the basis of the forecast risk, in order to successfully fight initially spread fires. In the real-time phase, optimal decision techniques must be used in order to determine the optimal composition of means that are sent to signalled fires.