RT info:eu-repo/semantics/doctoralThesis
T1 A non-intrusive and adaptive Digital Twin as enabling learning ecosystem for the development of predictive models in manufacturing environments
A1 García García, Álvaro
A2 Universidad de Valladolid. Escuela de Doctorado
K1 Desarrollo sostenible
K1 Industry 4.0
K1 Industria 4.0
K1 Digital Twin
K1 Gemelo Digital
K1 Human-machine interaction
K1 Interacción hombre-máquina
K1 Learning ecosystems
K1 Ecosistemas de aprendizaje
K1 33 Ciencias Tecnológicas
AB Recent unpredictable world economy challenges, such as the coronavirus pandemic and global energy crisis, have impacted the manufacturing industry, forcing production plants to reduce costs and improve productivity and sustainability. The demand for disruptive solutions and specialised workers under the Industry 4.0 paradigm has become an increasingly important digital priority for the manufacturing industry, which pushes technological upgrades towards building new cyber-physical ecosystems and supporting the skills improvement of the workforce. Despite the rapid adoption of next-generation Information Technologies, the accomplishment of this cyber-physical convergence remains an open issue in traditional manufacturing. In this way, the evolution of digital twins leveraged by progressive cyber-physical convergence has provided smart manufacturing systems with knowledge-generation ecosystems based on new models of collaboration between the workforce and industrial processes. However, industry will need to face the challenges of building and supporting new technical and digital infrastructures, while workers’ skills development eventually manages to include the increased complexity of industrial processes. Similarly, academia faces the challenges of providing technological research programs and experts in line with complex manufacturing life cycle processes. From the point of view falling between industry and academia, this PhD thesis is intended to reach a better understanding of human-machine learning opportunities offered by emerging Industry 4.0 digital twin ecosystems in manufacturing. To overcome knowledge acquisition barriers associated with traditional manufacturing, the proposed research activities have contributed to a set of incremental results obtained in industrial environments, which are summarised as follows: (i) understanding of the current enablers and challenges found in the digital twin cyber-physical convergence concerning human–machine collaborative ecosystems; (ii) original definition of Digital Twin Learning Ecosystem (DTLE) and presentation of its three-layer DTLE conceptual architecture; (iii) application of two case studies in traditional manufacturing to address both digital retrofitting and human-machine integration, without interfering in working conditions; (iv) development of a three-tier digital twin-based methodology and the knowledge modelling process focused on a non-intrusive cyber-physical twinned interaction between skilled workers and legacy systems, for building an adaptive DTLE in manufacturing; and (v) implementation and replication of a DTLE in two different traditional manufacturing Small and Medium-sized Enterprises (SMEs) under the actual human-machine work conditions.The results derived from this research culminated in a compendium of three publications. Based on these findings, the research priorities presented in this PhD thesis are considered a recognised basis in industry, which should help digital twins with the objective of progressive integration as a future learning ecosystem.
YR 2024
FD 2024
LK https://uvadoc.uva.es/handle/10324/69488
UL https://uvadoc.uva.es/handle/10324/69488
LA eng
NO Escuela de Doctorado
DS UVaDOC
RD 12-nov-2024