Antonios Korfiatis

• M. ARGYROPOULOU
• P. KOUROUTHANASIS
• D. ZISIS

This doctoral dissertation aims to develop and evaluate advanced demand forecasting techniques for the transportation and hospitality sectors with emphasis on tourism services. In a constantly changing tourism demand environment, accurate forecasting constitutes a critical factor for the sustainability and profitability of businesses in these sectors and the local communities that depend on them.
The dissertation focuses on developing an integrated demand forecasting framework that combines traditional statistical methods with modern machine learning techniques. Data from real cases of destinations (bed availability) and transportation modes (airports, passenger traffic) are being analyzed, taking into account multiple factors such as seasonality, economic indicators, weather conditions, and special events such as the COVID-19 pandemic.
Particular emphasis is placed on addressing the challenges arising from demand uncertainty and variability, as well as on developing models that can adapt to emergency situations and crises. The dissertation also examines the integration of real-time data and the utilization of big data technologies to improve forecasting accuracy (e.g., flight cancellations, etc.).
The results of the dissertation will contribute to improving decision-making in resource management, pricing, and strategic planning of businesses in the transportation and hospitality sectors, as well as to the overall management of the tourism service supply chain given the excessive demand observed in recent years. In parallel, a practical framework for implementing the proposed methods is being developed, which can be adapted to the specific needs of each business with particular emphasis on small enterprises that lack the necessary resources.