Dangerous goods transportation (DGT) represents technological and environmental risks for exposed populations, infrastructures and environment. Historical evidence has shown that road-accidents in DGT can lead to various potential consequences characterized by fatalities, injuries, evacuation, property damage, environmental degradation, and traffic disruption. Due to the importance of these products in everyday civil life activities and the increase in demand for these materials, developing tools for risk analysis and mitigation becomes a strategic goal in particular in those countries, like France, in which the majority of goods are transported by road. Based on the complexity of the dangerous goods transportation system DGTS and its related risk (factors that characterized risks are time dependent as traffic conditions, weather conditions, incident probability and population exposure), this analysis can only be made via simulation. This paper describes a generic approach to use agent-based modeling, an interesting approach to modeling systems comprised of autonomous and interacting agents, for risk analysis. It presents a novel generic model facet for representing risk analysis and fault tree propagation in an agent model, where the goal is to analyze the risk related to a system and to simulate its behavior in normal and degraded mode by using multi-agents systems. This approach is used to analyze the risks related to dangerous goods transportation and to minimize these risks by using agent-based model (identifying the best road that having the minimum risk level for transport).