Nishigaki, T., Schmöcker, J. D., Yamada, T., & Nakao, S. (2024). Using entropy maximisation for establishing city-wide touristic tour patterns. Applied Soft Computing, 154, 111316.

Comprehending the travel patterns of tourists is the foundation for sustainable urban tourism policies. We establish a tour-based entropy maximisation model to estimate the number of tourists partaking in specific tours. We define “tour” by the starting time, the places tourists visit and the order of the visits over the course of a day. Our main input data are the spatial-temporal distribution of tourists with respect to specific areas of the city. We formulate optimisation problems with and without additional knowledge of specific average tour characteristics such as tour starting times, the number of places visited, and the total tourist number. We find solutions that maximise entropy and minimise a penalty formulated as the sum of squared errors on constraints. Limitations are discussed due to non-convexity and the linearly independent condition not being met in some cases. We verify the efficacy of our method through a survey conducted by the Kyoto city government. It is shown that the errors on constraints are small, and our model successfully estimates the number of tourists partaking in each tour with a high degree of accuracy, provided appropriate constraints. Hence our methodology equips tourism planners with information to devise recommendations aimed at preventing over-tourism, thus fostering a more sustainable and enjoyable travel experience.