• clustering
• data-driven decision making
• performance optimization
• time series forecasting

This thesis presents a KPI Target Recommendation System, leveraging a multi-phase machine learning approach for a company in the hearing care retail sector. The system integrates time series forecasting models to predict performance metrics, clustering techniques to segment shops based on key attributes, and regression models to refine the prediction of the different Key Performance Indicators (KPI) targets. Combining these methods, the system delivers tailored recommendations that enhance decision-making and optimize retail performance.