A Data-driven Approach to Optimizing Travel Package Portfolios Based on Tourist Preferences

In this paper, a data-driven approach is proposed to optimize travel package portfolios based on tourist preferences. This approach consists of two stages. In the first stage, key attributes of travel packages are identified through the analysis of user-generated content, and then tourist preferences are analyzed by incorporating perceived trust and social learning behaviors of tourists. In the second stage, considering tourists’ choice behaviors, a bi-objective nonlinear mixed-integer optimization model for the travel package portfolio is constructed, and the NSGA-II method is applied to solve the model. Finally, a case study using Xi’an as a tourist destination illustrates the effectiveness of the proposed method. The results are discussed through comparative and sensitivity analyses. These findings can assist online travel agencies in gaining a comprehensive understanding of tourists’ needs and preferences, thereby improving the quality of travel packages and enhancing the tourist experience.