Siyuan Chen

Chalmers Tekniska Högskola, Kretslopp och vatten i Göteborgs stad, Mistra Inframaint

Siyuan holds a background in both computer science and industrial engineering. His research focuses on the utilization of artificial intelligence and digital twins to support smart maintenance. Siyuan’s work aims to develop an integrated manufacturing analytical platform designed to empower decision-making for industrial workers. His research interests encompass reinforcement learning, deep learning, generative AI, and digital twin technologies.

Biljetter och öppettider

Det är kostnadsfritt att gå på Underhåll, läs mer och registrera dig idag!
Hämta kostnadsfri entrébiljett
torsdag
Bridging the Divide: Aligning Academic Research with Industrial Reality in AI-Enhanced Digital Twins for Maintenance

While the convergence of Artificial Intelligence and Digital Twin technology is reshaping Industry 4.0, a significant gap persists between theoretical models and industrial implementation. This session presents a systematic analysis combining a review of 51 academic studies with in-depth interviews from industry practitioners to pinpoint critical disparities in scale, data complexity, and model robustness.

We will explore why industrial adoption is hindered by organizational barriers and data integration challenges, despite robust academic advancements in deep and reinforcement learning. Attendees will leave with a comprehensive five-layer framework and actionable recommendations which are designed to accelerate the practical realization of smart maintenance.

Spår:
Framtidens produktion
Lokal:
Teknikscenen

While the convergence of Artificial Intelligence and Digital Twin technology is reshaping Industry 4.0, a significant gap persists between theoretical models and industrial implementation. This session presents a systematic analysis combining a review of 51 academic studies with in-depth interviews from industry practitioners to pinpoint critical disparities in scale, data complexity, and model robustness.

We will explore why industrial adoption is hindered by organizational barriers and data integration challenges, despite robust academic advancements in deep and reinforcement learning. Attendees will leave with a comprehensive five-layer framework and actionable recommendations which are designed to accelerate the practical realization of smart maintenance.