1.04 Digital Services for Made in Italy: Digital-twins for predictive models and to support the lifecycle of Fashion products

REFERENCE SPOKE
OTHER SPOKES
PROJECT LEADER
Gustavo Marfia
START DATE
Gennaio 2023
END DATE
Dicembre 2025
PROPOSER
Alma Mater Studiorum – Università di Bologna
PARTNERS

Consiglio Nazionale delle Ricerche, Università degli Studi di Firenze, AEFFE S.p.A.

1.04 Digital Services for Made in Italy: Digital-twins for predictive models and to support the lifecycle of Fashion products

L’obiettivo del progetto è creare un digital twins che supporti il processo creativo durante le fasi di ideazione di abiti e accessori utilizzando paradigmi di intelligenza artificiale. Questo “gemello digitale” sarà quindi impiegato per pianificare, ottimizzare, tracciare il processo industriale necessario per produrre le sue repliche fisiche. Lo strumento conterrà allo stesso tempo una rappresentazione della sua storia, dalle origini creative alla produzione e alla sostenibilità, includendo anche informazioni di mercato e di vendita per stimare le tendenze e analizzare il punto di contatto tra marchio e consumatore. 

RISULTATI ATTESI

The project aims at adopting a digital twin-centric approach to the design, production, and commercialization of fashion products. To push such an approach and verify its feasibility within the considered domain of interest, the following results are expected within the project: (a) the definition of a digital twin abstraction which may include all of the information relevant during its design and production phases, on one side, and all of the information relevant for consumers, on the other. (b) analysis and verification of methodologies apt to transform 2d images into 3d meshes to be able to leverage existing fashion archives. (c) an analysis of the most advanced approaches, based on extended reality technologies and generative adversarial networks, apt to manipulate, transform 3d content and visualize it during creative processes. (d) verification of the information the digitization process of materials may reveal (e.g., quality, resistance, smoothness, etc.) and of how to include it within a digital twin. (e) the individuation of the technological infrastructure required to support the acquisition and prediction of logistical, material, waste, and trend information. (f) an analysis of the most suitable prediction methods that may employed with the aim of optimizing the industrial phase. (g) an analysis of how the e-commerce platforms may incorporate digital twins as the point of contact between a consumer and a brand. (h) the implementation digital twin prototypes that may serve as proof of concepts to let their advantages and eventual critical points emerge.