I. Background With the development of urbanization and motorization, road traffic accidents have caused serious casualties and social losses. A large number of traffic accidents also pose a significant challenge to the traffic police in maintaining road traffic safety. The project relies on the 'Shanghai World Expo Road Traffic Accident Analysis and Early Warning System', National Natural Science Foundation's 'Research on the Causation Analysis and Safety Evaluation Methods of Traffic Signal Control Intersection Accidents', 'Standardization of Road Names and Optimization of Accident Data Collection and Recording, as well as Demonstration Application', and other projects. Based on Shanghai, it has built and demonstrated the application of urban road traffic safety analysis and management technology based on accident data. The 'People's Daily' reported this in 2014. II. Main Content and Innovations 1. Established a semantic expression standard for illegal locations of traffic accidents, achieving intelligent positioning of traffic accidents. Innovation points: (1) Starting from the characteristics of traffic accident records, The Shanghai road network is divided into five major categories of road facilities and thirty types of accident locations, proposing a unified 'five-element' method for recording the illegal locations of accidents; (2) Based on the semantic representation of traffic accident locations in road facilities—accident sites—reference points, a scientific logical process is established, achieving intelligent positioning of traffic accidents on electronic maps. 2. A road traffic accident prediction model is established, and a system of influencing factors for traffic accidents is constructed. Innovation points: For the first time in China, a systematic prediction model for traffic accidents at signalized intersections, urban main roads, urban expressways, suburban main roads, and highway facilities is established. The causes of accidents are analyzed from three aspects: road geometric design, traffic control, and traffic operation, constructing a system of influencing factors for traffic accidents in different categories of facilities. 3. Standards and methods for identifying accident-prone facilities are established. Innovation points: The domestic first proposal of a safe and efficient spatial method was made, which systematically identifies accident-prone sections (intersections) on various types of road facilities in mega-cities, and proposes different evaluation criteria for different facilities based on the type and severity of accidents. This solves the problem of empirical identification of accident-prone sections (intersections). 4. A webGIS-based urban road traffic accident analysis and management system was developed. Innovation points: Based on WebGIS, complex road networks are classified by facilities and decomposed into intersections, road sections, and regions. The urban road traffic accident prediction model is integrated with the theoretical results of identifying accident-prone facilities, providing a professional solution for the analysis and management of traffic safety in mega-cities. III. Achievements 1. In 2010, the 'Shanghai City Road Traffic Accident Analysis and Early Warning System' was officially launched. It serves as the citywide and district/county road traffic accident analysis management platform for Shanghai. It has been used for 5 years so far. By 2014, there were 44 municipal and 60 district-level sections identified as accident-prone roads. The software copyright 'Road Traffic Accident Analysis System V1.0 Based on WebGIS' was obtained. 2. Since 2012, all traffic accidents and illegal records in Shanghai have been applied with standardized expressions and input controls, which has increased the accuracy of accident location from less than 50% to 98.2%. (1) Formed industry standard 'Shanghai Traffic Police Duty and Law Enforcement Location Name Special Expression Standard' (Shanghai Public Transport [Science and Technology] [2013] No. 4) (2) Obtained software copyright 'Road Traffic Accident Location Input and Real-time Positioning System Based on WebGIS' (3) Registered patent 'Method for Automatically Recording and Locating Semantic Expression Information of Traffic Accident Locations'. 3. Registered patent ' Method for processing traffic operation information of urban main roads based on floating car data. 4. Published 87 journal and conference papers, including 14 SCI journals, 24 EI journals, and 23 international conference papers. IV. Promotion In 2012, the technology was promoted in Guangzhou City through the implementation of 'Specialized Expression Specifications for Traffic Police Duty and Law Enforcement Locations', 'Traffic Management Road Basic Information Collection and Maintenance', and 'Comprehensive Analysis System for Illegal Traffic Accident Situations'. In 2013, Suzhou City learned and applied this technology. In 2015, Wuhan City carried out the work of 'Road Name Standard Compilation and Optimization of Accident Data Collection and Recording', and 'Traffic Analysis and Early Warning System' to promote this technology.