Spatial Analysis to Investigate the Relationship Between Tourism and Wellbeing in Italy

Abstract The level and variety of services offered by tourist destinations are intricately linked to the overall health and condition of its area. We would like to investigate the existence of a possible connection between tourism and the social, economic, and environmental well-being of a territory. The tourism industry can improve the general well-being of a specific area by promoting consumption, reducing the income gap, and improving infrastructures. However, the well-being of the territory through enhancing the specific features of the local context and its factors of excellence can also influence tourism. In this context, we applied Machine Learning methods to investigate the relationship between tourism and well-being in Italy. The analysis used Italian BES indicators at the provincial level, referred to a time window of 17 years (2004–2020). We developed a Machine Learning algorithm based on a hybrid (unsupervised and supervised) approach to study 51 well-being indexes and 9 tourism indicators. We found a close connection (80% of accuracy) between tourism and well-being. We also selected a group of tourism indicators that have a strong effect on this connection. Using eXplainable Artificial Intelligence (XAI) methods, we detected that tourism in low season periods ranks first for importance followed by the spread of farms business and urban green areas density. Our research suggests that improved social, economic, environmental, and health well-being can positively spill over the effect on tourism arrivals and revenues in the long period.