I developed a machine learning-based forecasting solution to predict the number of handling units required for a company’s internal logistics. The goal was to improve the accuracy of inbound logistics planning and reduce costs associated with under- or overestimating required resources. By analyzing historical data from SAP and implementing a Proof-of-Concept ML algorithm, I aimed to automate the prediction process and enhance resource allocation for the third-party logistics provider. While achieving the targeted 90% accuracy, the project also generated key insights and laid the groundwork for further improvements in logistics forecasting.