Road safety is one of the global issues in the world during travelling. The road accident causes critical effects in the lives of travellers. Thousands of people lose their lives every day, and hundreds of people are injured in a road accidents. Given the importance of the problem, identifying the causes of traffic accidents is the main objective to reduce traffic accidents. In this paper, we apply data mining algorithms and some statistical analysis of road traffic accidents are performed to find patterns and useful information. Different attributes like weather conditions, road surface, light condition, and severity of casualty are investigated. Apriori algorithm is used to discover association rules, and the ZeroR classifier is used to build the classification model. Our empirical results show that the proposed model could classify road traffic accidents with reasonable accuracy and find out association rules. Which help us to improve safety and prevention from accidents. As seen in the experiential setup, we find casualty severity on the basis of different attributes like road surface, weather condition, light condition, and type of vehicles. Association rules are applied to find out the best rules which prevent accidents and improve safety.