Alessandro Brusaferri received the Mechanical Engineering MSc degree (with specialization in automation and robotics), and the PhD in Information Technology (Computer science and engineering area) at Politecnico di Milano. Starting from 2004, he works at the Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing (formerly Institute of Industrial Technologies and Automation) of the National Research Council National Council of Italy (CNR-STIIMA), first as research fellow (2004-2011), and then as permanent staff researcher (2011- present). Currently, he serves as responsible of the CNR-STIIMA Intelligent Automation laboratory.

Summary of research activities: From 2004 to 2010, his research and development activity has been mainly dedicated to advanced industrial automation and distributed control systems (from IEC61131 to IEC61499 standard based systems), starting from the first design to the full shop-floor commissioning, through integrated methods and software tools.

From 2010 to 2017, he served as responsible, within the Machine and Manufacturing Control systems research group, for the development of enhanced control systems to enable the real-time optimization of energy intensive process industries (e.g., Iron and Steel, etc.). Chiefly, he worked at the realization of novel mathematical programming-based process supervision systems, implemented in SCADA platforms of real industrial plants, aimed to minimize electricity consumption, extend the participation to liberalized energy markets, and support the implementation of flexible demand response programs.

Currently, his research is mainly devoted to machine learning and deep learning approaches to multi-variate time series forecasting, automatic learning of behaviour models from data, and process optimization. His main interest is in the development of neural network based probabilistic forecasting methods (e.g., Bayesian deep learning, Mixture Density Networks, attention-based models, etc.) and tools (based e.g., on Tensorflow-probability, Pytorch-forecasting, etc.), to both reduce prediction error and convey prediction uncertainty in the model. He worked in several research and development projects including European initiatives (FP7, Horizon2020) and industrial contracts, assuming both technical and coordination roles.

Research interests: Machine learning, Deep learning, Time series, Probabilistic Forecasting, Industrial Automation, Advanced Control, IEC61499,

Degree: MSc degree in Mechanical Engineering (Automation and robotics area), PhD in Information Technology