Location Prediction of PTI Station Based on Machine Learning and GIS

Periodical technical inspection of vehicles plays a vital role in ensuring road safety. The selection of suitable locations for vehicle inspection stations, responsible for conducting the regular inspection, has long been a matter of significant concern for companies. This paper proposes a method for selecting optimal locations of vehicle inspection stations based on machine learning and real-time GIS electronic map data. Focusing on Beijing’s vehicle inspection station as a case study, this study establishes an evaluation model that encompasses factors like competitive relationships, business ecology, and population number indexes. Utilizing machine learning algorithms including Gradient Boosting DeciSion Trees (GBDT), Support Vector Machine (SVM), and K-Nearest Neighbors (KNN), this paper analyzes the unique environmental characteristics of the vehicle inspection station at chosen addresses. The results indicate that the GBDT algorithm is most suitable for station location prediction, yielding accurate results.