Palembang City is one of the big cities in Indonesia. Along with the increasing population and the increasing number of motorized vehicles, it will certainly have an impact on the increasing number of traffic accidents in the city of Palembang. In this study, the writer will determine the pattern of traffic accidents by using the fp-growth algorithm and using various variables. The variables that will be used consist of weather, time of incident, road geometry, profession, level of injury. This research is expected to be a reference for the police to be able to take anticipatory measures in order to reduce the number of traffic accidents in the Palembang City area. The fp-growth algorithm can be applied properly to determine the pattern of the causes of traffic accidents in the city of Palembang by using 2 minimum support of 40% and 50% and 2 minimum confidence of 70% and 90%. Based on the resulting rules, there are rules with the highest confidence value of 98% with these rules: When an accident occurs with a Side-Side accident type, the accident occurs in sunny weather conditions