Optimizing Taxi Route Planning Based on Taxi Trajectory Data Analysis

In daily life, taxis have become one of the most common and convenient ways of public transportation. With the advancement of positioning and mobility technologies, a large amount of taxi trajectory data has been collected, providing valuable data resources for urban planning, traffic management, and personalized route recommendations. However, these huge datasets also pose computational and processing challenges. This study uses the annual taxi trajectory data of Porto City obtained from the Kaggle platform, containing more than 1.7 million records, to study data query and analysis in a big data environment. We focus on comparing the efficiency and overhead of two spatial index structures, K-d tree and R-tree, in handling such large-scale datasets. Experimental results show that the K-d tree has a time-efficiency advantage in K-nearest neighbours query tasks, while the R-tree performs better in complex spatial query tasks. These findings provide important references for taxi route planning and other big data applications, especially in scenarios requiring efficient and accurate data retrieval.