This work presents a digital-twin-based framework focused on orchestrating human-centered processes toward Industry 5.0. By including workers and their digital replicas in the loop of the digital twin, the proposed framework extends the traditional model of the factory’s digital twin, which instead does not adequately consider the human component. The overall goal of the authors is to provide a reference architecture to manufacturing companies for a digital-twin-based platform that promotes harmonization and orchestration between humans and (physical and virtual) machines through the monitoring, simulation, and optimization of their interactions. In addition, the platform enhances the interactions of the stakeholders with the digital twin, considering that the latter cannot always be fully autonomous, and it can require human intervention. The paper also presents an implemented scenario adhering to the proposed framework’s specifications, which is also validated with a real case study set in a factory plant that produces wooden furniture, thus demonstrating the validity of the overall proposed approach.

This paper analyzes the potential of Semantic Web technologies to support innovation in industrial scenarios. The study focuses in particular on RDF stores, the software components dedicated to the storage, representation and retrieval of semantic information. Starting from a literature review, a qualitative analysis is carried out considering a set of these systems. RDF stores are evaluated to find the implementations that are best suited to play the role of backbone in a software architecture sharing information between the software tools adopted by the various stakeholders. This can be achieved if the architecture overcomes the problems deriving from the lack of integration between the involved software applications, providing in this way an integrated view on relevant engineering knowledge.

The problem of achieving a good maintenance plan is well-known in the modern industry. One of the most promising approaches is predictive maintenance, which schedules interventions based on predictions made by collecting and analyzing data from the process. However, to the best of the authors’ knowledge, this approach is still not widespread and known enough, and particularly, the real-case scenarios of its application appear not exhaustive. To contribute to fill this gap, this work proposes a digital twin (DT), which performs a predictive maintenance approach for a conveyor belt within a real-case scenario with the overall goal of predicting faults during normal belt operations. Specifically, the core of the implemented DT is a model that analyzes the data collected by various sensors distributed along the conveyor belt. In turn, this model exploits a machine learning-based algorithm that predicts the insurgence of faults. The tests of the developed solution, conducted within a real scenario, demonstrated good precision and accuracy in identifying the fault status and also in a time deemed acceptable for the involved stakeholders.

Data integration is one of the most crucial challenges for current manufacturing companies. Indeed, while the extended use of software tools generate more and more data about products, processes, and production resources, still this huge amount of data is represented using different formats and non-aligned structures. This issue is worsened by the fact that data can be found scattered in not linked databases and hosted in mutually incompatible systems. The growing need to access these data and the knowledge that they encapsulate on a global view from different perspectives is addressed by various approaches in order to support the integration between involved systems. This paper deals with the applicability of Semantic Web technologies in industrial context to enhance semantic interoperability. In particular, it proposes a systematic approach to support the development of a semantic model, focusing on the combination of two main critical aspects: the reuse of existing reference models and the semantic migration of the existing legacy models. Even both the aspects have been separately studied in previous research works and even their overlap is recurring during the stage of ontology development, to the best of our knowledge the literature is missing studies covering the synergistic and automatic combination of these two aspects within a holistic approach. The application possibilities of the approach are also investigated within a real case study from a high-precision mould-making company; thus demonstrating its feasibility for use in complex manufacturing contexts.