The study aims to identify factors affecting traffic accident concentration using spatial data science techniques. It analyzes traffic accident data from Barcelona, Spain, and other relevant datasets to understand the impact of various factors such as age, type of vehicle, trip purpose, weather conditions, human mobility, and road characteristics on accidents. The analysis reveals that accidents are highly influenced by space-time patterns, with hotspots concentrated in specific areas depending on the day and month. The study also identifies six spatial outliers (High-Low) locations with large roundabouts or complex road systems, which correspond to higher accident rates. A predictive model is trained using Regression Kriging with a Random Forest regressor to predict annual accidents per cell, highlighting the importance of factors such as traffic variation, human mobility, and road characteristics. The findings have significant implications for cities, logistics companies, insurance companies, and public resources management, enabling better organization of traffic control systems, route optimization, and safer driving experiences.