Geographia Technica, Vol 20, Issue 2, 2025, pp. 179-189
FORECASTING REGIONAL TOURISM FLOWS IN ALBANIA USING ARIMA AND SPATIAL ANALYSIS: A DATA-DRIVEN APPROACH BASED ON NATIONAL ACCOMMODATION STATISTICS
Medjon HYSENAJ
, Ditmira TAHIRI 
ABSTRACT: Geographic Information Systems (GIS) have become essential tools for tourism planning and management, enabling spatial analysis and advanced data visualization. This study adopts a data-driven approach to forecast and interpret regional tourism flows in Albania by integrating GIS techniques with statistical time series modelling. The spatial component involves mapping tourism activity across three macro-regions North, Centre, and South while distinguishing between coastal and non-coastal zones using official data from the Albanian Institute of Statistics (INSTAT). To complement the spatial analysis, the study introduces a temporal dimension using a matrix-style visualization that highlights seasonal tourism trends throughout 2024, and applies the ARIMA (AutoRegressive Integrated Moving Average) model to generate regional forecasts for 2025. This combination of spatial and temporal methods enables a more comprehensive understanding of tourism dynamics, revealing both geographic concentration and predictive seasonal patterns. The findings underscore the dominance of the Central Region, the growing potential of the North, and the strong seasonal pressure on coastal areas during peak months. By integrating GIS, heatmaps, and ARIMA forecasts, the study provides a strategic framework for sustainable tourism development and more informed regional policy interventions.
Keywords: GIS, ARIMA, tourism forecasting, spatial analysis, heatmap, seasonality modelling