Enhancing Educational and Tourism Strategies at Heritage Sites: Visitor Flow Prediction Using ARIMA and Autoregressive Models

Abstract

This study focuses on the use of ARIMA and Autoregressive (AR) models to predict visitor flow to Civil War shelters in Alicante, highlighting seasonal patterns and differences among various visitor groups, with an enriching approach toward educational and tourism applications. Through a retrospective longitudinal design covering from August 2023 to January 2024, it analyzes the time series of visits, differentiating between the general public and school groups, as well as examining geographical demand. The research emphasizes the effectiveness and simplicity of the ARIMA(0, 0, 0) model with logarithmic transformation in modeling time series, while the AR(6) model proves indispensable for capturing short-term temporal dependencies. Despite the usefulness of these forecasts for future planning, the existence of uncertainties highlights the importance of adopting flexible management approaches and incorporating additional variables to refine predictions. This approach not only improves the management of visitor flows but also significantly contributes to the creation of more effective educational and tourism strategies, promoting the sustainability and appreciation of cultural heritage.