Using demand to prevent ageing and depreciation

Revenue management is a broad field of models and methods that is growing in popularity. The goal of all these models is optimizing the expected revenues. In this thesis, we focus on two specific techniques within revenue management. Both techniques focus on forecasting future events based on uncertain information from history: Demand time series forecasting and dynamic pricing. The company Tech Data has asked to think of new ways to increase their revenue and inventory model to decrease depreciation costs and prevent aging. We discuss several time series forecasting methods and use them to determine the optimal method for Tech Data. Furthermore, we introduce some additional techniques such as hierarchical forecasting and regression models. Secondly, we use dynamic pricing to researches methods that model the correlation between demand and price setting. The goal is to find a model that optimizes revenues and could be used by Tech Data. We combine a theoretical basis with numerical experiments on live data from Tech Data.