Abstract
Tourist centers need help managing visitor numbers for specific events or seasons. This thesis presents a decision support system for tourist centers called Predictive Visitor Management Application (PVMA). Based on previous years' statistical data, the system aims to assist the staff of tourist centers in anticipating the number of visitors for specific events or seasons. The system identifies trends and patterns influencing visitor numbers by analyzing historical statistical data and corresponding Google Trends data to consider tourists' interests and preferences. Furthermore, climate data is integrated to account for the influence of weather on visitor patterns, ensuring a more comprehensive prediction model. The system design of PVMA aims to support flexibility and extensibility at both architectural and module levels facilitating the integration of new functionalities and data sources and ensuring adaptability to changing requirements. The high- level design adopts the microservice architecture to address the data-assembling, prediction strategies, and result presentation. This high-level design enables the flexibility to integrate new services and alternative data sources to accommodate varying requirements. The system incorporates the strategy pattern at the module level design, enabling efficient management of prediction algorithms and strategies. This design choice eases the addition of new algorithms and the ability to update or remove existing ones as needed. In obtaining predictions for the visitor influx, time series algorithms such as ARIMA are employed for training and generating forecasts, enabling it to discern recurring patterns and seasonality in visitor behavior. The PVMA can identify long-term trends and short-term fluctuations by considering the day of the week, month, or year from the historical statical data provided. This level of analysis helps the system forecast the visitor numbers to the tourist center staff, enabling them to be prepared for upcoming events. This study designs and develops a microservice-oriented Predictive Visitor Management Application for tourist centers to forecast and prepare for visitor influxes. The flexibility and extensibility of the system ensure that the PVMA can be tailored to the specific requirements and characteristics of each tourist center.