This paper presents an example of integration between open source Geographical Information System (GIS) and Agent Based Model (ABM) in order to better simulate fishing activities on Iroise Sea (Brittany, France). This work is linked with the SIMARIS project: a simulation prototype that integrates multi-source and multi-scale spatio-temporal constraints as forcing variable in order to assess the intensity and the variability of marine activities. A pre-processing step, executed in batch in GRASS GIS, aims to calculate data for initialization and simulation step, then the Agent Based simulation is launched (in batch) on GAMA platform. All these operations are scheduled in a Python script to perform pre-processing and simulation. The work presents an example of integration from a geographical point of view. The technical improvements are detailed and the potentials of such integrated solution are discussed.