Mémoire-projet

Abstract

Sales forecasting is a complex but essential exercise for any company. This process helps to optimise supply chain management, resources allocation, production planning and, most importantly, to improve customer satisfaction by responding more effectively to demand. In recent years, artificial intelligence (AI) has made significant advances, opening the way to new business applications. A number of forecasting solutions based on AI are now available on the market. But are these tools really based on advanced technologies, or are they simply riding the AI trend? What are the concrete mechanisms behind them? Through the case study of Sauermann Group, this thesis seeks to assess the impact of AI on two key aspects: the accuracy of sales forecasts and the operational gains it can generate. To achieve this, several forecasting models, including one based on machine learning, have been compared using performance indicators and graphical analyses to help interpret the results. The second part of the analysis is based on a simulation comparing the time spent on the forecasting process in its current version versus a version incorporating a machine learning tool. This comparison was carried out for both the S&OP department and the regional sales managers, who are key stakeholders in the process. The aim of this work is to demonstrate the extent to which AI-based forecasting models can be an alternative or a complement to traditional approaches, while adding real value to the company's decision-making process.