Mathematical models for selling process optimization

The work of the Thesis has been pursued in collaboration with an important company operating in the tourist sector. The followed projects in the work can be seen as belonging to the Destination Management branch, that is the study and the implementation of actions aimed at better managing the company offer related to touristic experiences broadly. In particular, the first project has been related to Revenue Forecasting and has dealt with the definition of a methodology, based on mathematical and statistical techniques, aimed at forecasting the revenue streams linked to specific items of a company; the second project, Destination Discovery, instead aimed at the high-level analysis of tourism opportunities in different geographical areas, researching and evaluating new possibilities for the company related to tourist interest. In the work, some preliminaries about what a forecast is will be provided and the many techniques aimed at accomplishing the task, giving a general theoretical framework for the topic will be discussed. Then some details about data and the set of more practical operations needed in order to extract information from them in a numerical manner, eventually building a forecasting model on it will be discussed. Later the application of the techniques previously introduced to the different projects will be discussed; for each one of them, the methodologies followed and the analyzes carried out will be provided, as well as the obtained results. Finally an analysis of what has globally been done in the work along with different comments, the obtained results and some possible future work and improvements will conclude the work.