Big data has become a fundamental game changer in the industrial and service sector over the last few years, but only recently there has been a significant shift of focus from the hype surrounding it to finding real value in its use. However, a number of barriers have to be reduced related to mastering the complex and rapidly evolving tools and techniques, introducing substantial delays and costs in product/process design, deployment, test, and refinement.
These models, data-driven, simulation-based, visual etc., enable the description of the system itself and its dynamics (descriptive or interpretative models), the prediction of its evolution (predictive models), and the optimization of its operation, management and maintenance (prescriptive models). They are used to detect and diagnose anomalies, to determine an optimal set of actions that maximize key performance metrics, to effectively and efficiently enforce on-line quality management of production processes under latency and reliability constraints, and to provide predictions for strategic planning to help companies, especially SMEs, to significantly improve their profitability through digitalization, as well as to open up new opportunities for them for the creation of new services and business models.
IPR / Licence
Rely on widely spread open source solutions and tools in the field, namely Python as a programming language and a diverse set of libraries employed by many practitioners (both in the research as well as industry communities), such as TensorFlow, scikit-learn, numpy, pandas, and so on. In addition, all developed services are implemented in the form of dockerized containers.
Mobile Middleware Group Research Group website: https://www.middleware.unibo.it/