This webinar was on the general workflow of predictive maintenance models in order to understand the key aspects and the critical points aimed at obtaining accurate and robust algorithms. The workflow was employed for a real case study – a flow pack machine. Using sensors and synthetic data, the webinar explored importing, pre-processing, and labelling data, as well as selecting features (A.K.A healthy/faulty condition indicators) aimed at training and comparing multiple machine learning models. The lecture also focused on how MATLAB and Simulink can offer an integrated platform to design a digital twin and implement machine learning algorithms. The MATLAB environment provides dedicated capabilities to help the user to face Predictive Maintenance through a systematic approach, as well as a quick and easy interactive experience. The talk was given by Dr Marco Rossi is an EDU Customer Success Engineer at MathWorks, Italy.