ABSTRACT
Sustainable production aims at creating products from processes that minimize environmental impact, energy consumption and natural resources. Customers and markets are ever more leaning towards digital, custom, and flexible solutions with lower environmental impact. Hence, Industry 4.0 (I4.0) solutions are increasingly including social and environmental sustainability aspects. We focus on the realization of an infrastructure integrating industrially relevant application modules by combining system reconfigurability and artificial intelligence, towards sustainable production. To meet the final goal of sustainable production, we address four challenges considering flexibility and sustainability in production in a holistic way: (1) developing infrastructural and methodological tools to support companies to explore the potential of I4.0 towards sustainable production; (2) managing the configurability and customization possibilities of products; (3) effectively handling the flexibility provided by a production system with rapid reconfiguration capabilities; (4) integrating hardware and software flexibility by using reconfigurable robotics and machine learning methods. By developing and connecting different application modules, we obtain a physical demonstrator which represents on the one hand an exemplary scenario of reconfigurable and flexible production system; on the other, it enables new research activities and insights with a see, touch & feel approach for industrial and research realities.
ABSTRACT
Smart Agriculture (SA) is an evolution of Precision Farming (PF). It has technological basis very close to the paradigms of Industry 4.0 (Ind-4.0), so that it is also often referred to as Agriculture 4.0. After the proposal of a brief historical examination that provides a conceptual frame to the above terms, the common aspects of SA and Ind-4.0 are analyzed. These are primarily to be found in the cognitive approaches of Knowledge Management 4.0 (KM4.0, the actual theoretical basis of Ind-4.0), which underlines the need to use Integrated Information Systems (IIS) to manage all the activity areas of any production system. Based upon an infological approach, "raw data" becomes "information" only when useful to (or actually used in) a decision-making process. Thus, an IIS must be always designed according to such a view, and KM4.0 conditions the way of collecting and processing data on farms, together with the "information precision" by which the production system is managed. Such precision needs, on their turn, depend on the hierarchical level and the "Macrodomain of Prevailing Interest" (MPI) related to each decision, where the latter identifies a predominant viewpoint through which a system can be analyzed according to a prevailing purpose. Four main MPIs are here proposed: (1) physical and chemical, (2) biological and ecological, (3) productive and hierarchical, and (4) economic and social. In each MPI, the quality of the knowledge depends on the cognitive level and the maturity of the methodological approaches there achieved. The reliability of information tends to decrease from the first to the fourth MPI; lower the reliability, larger the tolerance margins that a measurement systems must ensure. Some practical examples are then discussed, taking into account some IIS-monitoring solutions of increasing complexity in relation to information integration needs and related data fusion approaches. The analysis concludes with the proposal of new operational indications for the verification and certification of the reliability of the information on the entire decision-making chain.
