Road accidents cause enormous human and financial losses all around the world. This research presents an Accident Prediction System (APS) that predicts the probability of traffic accidents by applying the Decision Trees algorithm. To build decision trees for prediction, the APS incorporates weather, traffic patterns, road infrastructure, and past accident data. A dataset containing numerous accident-related characteristics, such as location, vehicle speed, road type, time of day, and environmental factors, is used to train the system by applying the decision trees algorithm. Performance evaluation of this system shows promising accuracy in accident prediction through rigorous evaluation and validation processes. In addition to identifying high-risk locations that are prone to accidents, the APS also helps develop preventative strategies for traffic control and accident avoidance. By giving stakeholders, such as traffic authorities, urban planners, and law enforcement agencies, a trustworthy forecasting tool to lessen the frequency and severity of traffic incidents, this research helps to improve road safety.