AI-Driven Digital Twins in Action: From Cloud Native Systems to Advanced Industrial Applications
Date and time: 12 December 2024, 13:00-14:00 CET
Speaker: Michel Gokan Khan, KTH
Title: AI-Driven Digital Twins in Action: From Cloud Native Systems to Advanced Industrial Applications
Where: Digital Futures hub, Osquars Backe 5, floor 2 at KTH main campus OR Zoom
Directions: https://www.digitalfutures.kth.se/contact/how-to-get-here/
OR
Zoom: https://kth-se.zoom.us/j/69560887455
Host: Mario Romero or Jan Kronqvist
Abstract: Digital twins, while widely applicable, vary significantly in their design and purpose depending on the context. In this talk, I will present recent work demonstrating how AI can potentially automate the creation of digital twins and how these distinct applications serve various yet unique purposes and employ specialized technologies. I will discuss how industrial digital twins, powered by sensor data and advanced techniques like computer vision, 3D reconstruction, SLAM (Simultaneous Localization and Mapping), object detection, and 3D Gaussian Splatting, can provide immersive visualization and actionable insights for front-line operators and quality assurance teams, capturing spatial and procedural knowledge to enhance system performance and support effective skill transfer.
I will also discuss our contributions focused on generating digital twins of cloud native systems, highlighting how they optimize microservice deployments, enhance resource allocation, and support seamless scalability. I will showcase recent projects in these two distinct contexts, address key challenges, and explore future directions to illustrate how AI-driven digital twins enhance operational agility and efficiency across different domains.
Bio: Michel Gokan Khan earned his PhD in Computer Science from Karlstad University, Sweden. His dissertation focused on leveraging machine learning for the optimization of microservice chains and automated digital twinning of cloud native systems. Michel is currently a Digital Futures industrial postdoctoral fellow at KTH Royal Institute of Technology, working on a project in collaboration with AstraZeneca, leveraging machine learning and computer vision for AI-driven digital twinning of production lines and to optimize Overall Equipment Effectiveness (OEE), aligning with Industry 5.0 initiatives.
His research spans machine learning, cloud/edge computing, computer vision, and ML-driven optimization. He also has extensive experience in software architecture, large-scale SaaS systems and has held leadership roles as a CTO, tech lead, and engineering manager in various companies and has received awards for his contributions, including IEEE Best Paper Awards at NetSoft ’20 and NFV-SDN ’18, recognition as Sweden’s IEEE Young Professional of the Month in 2019, and a nomination in the Swedish Game Awards 2019, among others.