Resumen: We present a novel method, within the realm of data-driven computational mechanics, to obtain reliable and thermodynamically sound simulation from experimental data. We thus avoid the need to fit any phenomenological model in the construction of the simulation model. This kind of techniques opens unprecedented possibilities in the framework of data-driven application systems and, particularly, in the paradigm of industry 4.0. Idioma: Inglés DOI: 10.1063/1.5034931 Año: 2018 Publicado en: AIP Conference Proceedings 1960 (2018), 090005 [5 pp.] ISSN: 0094-243X Factor impacto SCIMAGO: 0.182 - Physics and Astronomy (miscellaneous)