Biography:

Andrea Matta is Full Professor of Manufacturing and Production Systems at Department of Mechanical Engineering of Politecnico di Milano and Guest Professor at Shanghai Jiao Tong University. He graduated in Industrial Engineering at Politecnico di Milano where he develops his teaching and research activities since 1998. He was Distinguished Professor at the School of Mechanical Engineering of Shanghai Jiao Tong University from 2014 to 2016. He has been visiting professor at Ecole Centrale Paris (France), University of California at Berkeley (USA), and Tongji University (China).
He is scientific responsible of the Research Area Design and Management of Manufacturing Systems at MUSP (Laboratory for Machine Tools and Production Systems). His research area includes analysis, design and management of manufacturing and health care systems. He has published 130+ scientific papers on international and national journals/conference proceedings.
He is Editor in Chief of Flexible Services and Manufacturing Journal since 2017, editorial board member of OR Spectrum journal and IEEE Robotics and Automation Letters journal. He is Co-Chair of the technical committee IEE RAS Sustainable Production Automation. He is member of scientific committee in several international conferences. He was awarded with the Shanghai One Thousand Talent and Eastern Scholar in 2013. He has been cited among the Leading Scholars in Production Research by International Journal of Production Research journal.

Speech title: Automated discovery and generation of digital twins for manufacturing systems


Abstract:
With the coming of the Industry 4.0 wave, digital representations of manufacturing systems have been promoted from marginal to central. Digital twins are not simply conceived as simulation models of their physical counterpart for off-line what-if analysis, differently they are developed as self-adaptable and empowered decision-makers timely aligned with the dynamics of the real system. Enriched from these new features, digital twins are widely recognized as the key enablers for the implementation of a smart manufacturing paradigm. Despite this new role, there are significant barriers to the adoption of the digital twin concept in industrial applications. In fact, the creation and update of digital twins is still a challenge because of the high skills required to use the simulation applications available in the market, the long development times, and their difficult integration with optimization and artificial intelligent packages. The frequent changes manufacturing systems encounter in their life cycle boost these issues. This talk describes a data-driven approach for generating multi-fidelity digital twins of manufacturing systems from data acquired from sensors. The approach is based on process mining techniques, tracing the objects flowing in a discrete event system allows us to discover the system topology and to automatically generate the digital twin. Being based on traces containing sensor-data, the approach enables multiple actors to jointly construct digital twins of complex systems. The approach is not limited to manufacturing, but extendable to other sectors to discover complex flows along the supply chain and back to reuse/remanufacturing/recycling in circular economies. Test cases and applications from the automotive industry are presented with the intention of quantifying the benefits of the approach, understanding its actual limitations, and discuss future perspectives.