This research was done to predict future forest fires using the multi-criteria decision-making (MCDM) method in District Three of Neka-Zalemroud Forests. We used a fire risk model integrated with the MCDM method in a GIS framework to map the forest fire risk in the study area. Factors included four major criteria (topographic, biologic, climatic, and human factors) and 17 sub-criteria. Data of these parameters were collected and organized in the GIS framework to provide the factor maps. Then the major criteria maps (using weighted overlay of sub-criteria maps of each criterion) and the fire risk map (using weighted overlay of four major criteria maps) were provided using the fire risk model. The actual fire data in the study area was used for cross checking. Results of this study showed the high-risk regions in fire risk map accordance with the actual fires. It can prove the high accuracy of the MCDM method and the used model to predict future fires in Hyrcanian Forests of Iran.