Tourism transportation research has become increasingly prominent due to the rapid expansion of Internet technology and the overwhelming availability of information. This abundance poses challenges in delivering tailored services that meet diverse user preferences. Consequently, personalized tourism transportation has emerged as a predominant trend. This paper proposes a mathematical model for personalized travel planning by integrating mainstream tourism service analysis with multi-source traffic data. The proposed approach introduces a two-stage spatiotemporal network solution algorithm. In the initial stage, utilizing a specified set of travel attractions, it employs a shortest path algorithm to determine an approximate optimal route that aligns with traveler preferences and supports multiple transportation modes. The second stage utilizes the spatiotemporal network to facilitate daily travel planning across multiple attractions. This algorithm effectively addresses path planning challenges and simplifies route planning with time constraints, offering valuable insights for future recommendations in personalized travel planning.

Tourism transportation, personalized travel planning, spatiotemporal networks, shortest path algorithm, multi-source traffic data.