It is increasingly recognized that fisher‟s behavior should be taken into account to understand and predict fishery dynamics, in particular in response to management. ISIS-Fish is a spatially and seasonally explicit modeling framework especially designed to couple populations and fleets dynamics and explore the impact of various management measures on mixed fisheries. It has already been set up using a static fishing effort allocation (corresponding to an average historical pattern) between the various métiers to simulate the pelagic fishery in the Bay of Biscay. We present here the integration of a fleet dynamics model. This model is derived from a Random Utility Model simulating métiers choice using as explanatory variables the past value per unit effort, the average percentage of effort spent in the different métiers and the fuel costs associated to each métier. Simulation results while applying the dynamic effort allocation are compared with observed effort allocation over the period 2000-2004, period where no management constraints on the fishery were observed, and the period 2005-2008, period where the anchovy fishery was closed. The simulated effort allocation fits observations over the period 2000-2004 for some métiers, but not for the most variable ones. During the period 2005-2008, the dynamic effort allocation enables us to reproduce the effort reallocation from anchovy métier to métiers targeting other species. We also reproduce very well the small period where anchovy fishery was reopened in 2005.