Application for recommending accommodation prices

The thesis examines the area of revenue management and pricing for accommodation facilities. Setting the right prices critically affects the profitability of a business, but the multitude of factors affecting demand make it a very challenging discipline. Previo s.r.o., a company specializing in the creation of software solutions for small and medium sized businesses, expressed an interest in developing a software solution that would enable them to obtain pricing recommendations based on consideration of appropriate aspects. The main objective of the thesis is to design and implement a custom application used for recommending room rates of accommodation establishments. The first sub-objective is to select suitable parameters for price optimization, the second sub-objective is to select a suitable algorithm for price recommendation and the third sub-objective is to create an API for communication with hotel systems. To meet the main objective of the thesis, MMSP-AV methodology was used to design and implement the application using Python programming language, Django framework and XGBoost machine learning algorithm. Suitable parameters were selected in consultation with experts in the field of pricing for accommodation facilities. The appropriate algorithm was selected based on a systematic literature search and the required API was created during implementation. User and functional testing was also carried out during which the usability of the developed solution was verified. The result of the thesis is a working prototype of the application, which, based on the input parameter values for individual room types of accommodation facilities, appropriately suggests price recommendations that can be used as a basis for price optimization. The source code of the application is freely available under the MIT license and is hosted on the GitLab repository.