色中色

Topic 

Leveraging AI/ML and DT to solve complex problems in different applications domains

Abstract

Digital Twins (DT) and Artificial Intelligence (AI)/Machine Learning (ML) have emerged over the last decade as key technologies at the heart of the digital transformation to enhance decision-making and improve/optimize different aspects of systems and processes. A Digital Twin (DT) is a digital representation of an actual object, system, or process, referred to as the Actual Twin (AT), that is dynamically and continually updated with system data, and provides a set of services related to the AT. It is used to monitor, improve, and optimize various aspects of ATs.

DTs are now used in a broad range of application domains, including aerospace, agriculture, automotive, construction, earth monitoring, healthcare, high-tech, robotics, smart buildings, smart cities, and telecom. Conversely, AI and ML empower organizations to leverage vast amounts of data to support better decision-making, enhanced efficiency, and drive innovation. These techniques can be used to analyze large datasets, both structured and unstructured, to discover patterns, trends, and insights that humans may overlook. Moreover, AI and ML enable real-time data processing, providing insights that support timely decision-making and operational adjustments. However, while AI and ML provide advanced capabilities, they do not inherently solve problems on their own. They require high-quality data and effective mechanisms to ensure that AI and ML components significantly enhance specific aspects of a system or process.

In this masterclass, we will investigate how DT and AI/ML technologies can be combined to build more effective solutions to solve complex problems in different applications domains. In particular, we will investigate how AI/ML can be used to implement advanced DT services. Also, this masterclass aims at leveraging the experience/expertise of the participants in AI/ML to propose new research directions for DT engineering.

TARGETED AUDIENCE

Researchers, PhD's, Postdocs, and (advanced) MSc students

FORMAT

This masterclass will take the form of a 2-day (10h00 to 16h00) workshop  that will include presentations of core techniques and concepts, breakout work sessions, and group discussions. 

Guiding questions:

• What are the main challenges related to use of AI/ML techniques in DT?
• What type of services can AI/ML provide to DT that other types of conventional technologies cannot provide?
• How would such a service benefit stakeholders, including user of DT/service, developers/maintainers of a system, domain experts?
• What is the role of AI/ML in DT engineering?
• What AI/ML techniques are best suited for DTs?
• Can AI-Agents bring DTs to the next level?
• What are the main challenges in using AI/ML in DTs?

Proposed agenda

Day1

• Part 1: Introduction to DTs 鈥擳his part will include presentation and discussion of the main DT concepts
  • Definition of DT and examples of DTs in different application domains
  • DT architecture and its different layers/components
  • Group discussion driven by the following questions:
    • How can DT be used to solve main issues faced by our society in the context of the digital transformation? 
    • What are the main challenges, benefits, and costs related to the use of DTs?

• Part 2: AI/ML
  • Introduction to use of AI/ML in DT 鈥擳his part will include presentation and discussion on the use of AI/ML in DT
  • Breakout session: Group discussion on the main challenges, benefits, and costs related to the use of AI/ML in the industry to solve complex problems
  • Plenary: Each group to present the outcome of their group discussion

Day 2

ATTENTION: This master class will be held over 2 days. Please register separately for both days. Go here to Day 2 for more information and registration. 

About the speaker

Professor Francis Bordeleau is at Ecole de technologie superieure (ETS) in Montreal where he leads the Kaloom-TELUS ETS Industrial Research Chair in DevOps. His current research focuses on the systematic improvement of industrial DevOps processes and on the engineering of digital twins for different application domains. His research aims at leveraging the best practices of software engineering, DevOps, model-based engineering, and AI/ML to improve various aspects of software processes and products. He has over 25 years of experience researching, working, consulting, and collaborating with companies worldwide in software engineering, software process improvement, and Model-Based Engineering (MBE). Prior to joining ETS, he was Product Manager of Software Development at Ericsson (2013-2017), founder and CEO of Zeligsoft (2003-2013), and assistant professor at Carleton University (1997-2006). Francis has been part of the organizing and program committee of many international conferences and workshops.

Your host

Loek Cleophas, assistant Professor in Engineering of Software Intensive Systems of the department of M&CS, will host Professor Franics Bordeleau of Software Engineering, Kaloom-TELUS 脡TS Industrial Research Chair in DevOps.

Registration is required but free of charge.
While attending only one of the two days is allowed, it is recommended that participants attend both.