The main objective of this project is to design and implement a GIS-based model to determine the high potential areas for hydrocarbon reserves for further exploration work, including seismic surveys and exploration drilling. In this project, a data-driven approach, evidential belief function, were used to implement the model. A case study in Ahvaz/Khuzestan province, south-west of Iran, is proposed to assess the feasibility of this modeling technique. Factors associated with hydrocarbon resources were assembled in a spatial database and a hydrocarbon resource potential map was created using the evidential belief function model. The method quantifies the spatial relationship between input data and the petroleum resources. This allows the exploration manager to identify those data themes that are the best predictors of petroleum prospects.