---

Road transportation by way of automobiles is a very convenient means of transportation. Today, the most detrimental consequence of developing transportation systems in a country is traffic accident that places a huge financial burden on society. This paper investigates the role of information systems in transportation safety that leads to improved planning and operation of the transportation system through the application of new technologies. Current methods for identification of segments of roads with high potential of accident are based on statistical approaches. Since there are not accident records for newly built roads, these methods cannot be used for regional roads that are recently built. This paper presents a GIS based Neuro-Fuzzy modeling for identification of road hazardous zones. The results of proposed approach are compared with statistical methods. It is shown that this method is a cheaper but at the same time robust means of analyzing the level of hazard associated with each road segment under consideration, specially when data are uncertain and incomplete.