We are performing the digital transition of our industrial and service activities, we are living the 4th industrial revolution, we are building the new World where  the digital, physical and human dimensions are strongly interrelated in socio-cyber-physical systems. The internet of things, big data, the industrial internet, are changing the way we design, manufacture, deliver products and services. For the sustainability of these transformations, the relevant knowledge, information and data must be integrated within model-based and data-driven approaches for the assessment and prediction of system and process behaviors, aiming at improved safe and reliable design, operation, management and maintenance.
There is a plethora of methods available and many potential applications for them: yet, there are still a number of critical problems which impede their full exploitation in practice. In this lecture, I frame what Machine Learning does, both theoretically and by practical examples, and offer some reflections on the challenges still remaining for a full deployment in practice.