AI for Time Series Analysis: From Anomaly Detection and Predictive Maintenance to Foundation Models

Time series data is everywhere, from industrial machines to energy systems. Discover how AI turns sensor measurements into forecasts, anomaly alerts, and early warnings of equipment failures, and explore what foundation models can (and cannot) deliver.

Industrial systems continuously generate time-stamped measurements, from temperatures, vibration, and electrical signals to production and energy consumption. These data offer opportunities to detect abnormal behavior, anticipate failures, and forecast future conditions. Yet real-world time series are often noisy, high-dimensional, non-stationary, and sparsely labeled, making reliable analysis far from straightforward.

This webinar introduces the core challenges and AI approaches for time series analysis, moving from classical statistical techniques to deep learning, generative models, and Time Series Foundation Models (TSFMs). We will distinguish anomaly detection, forecasting, and predictive maintenance, and explain how multivariate models learn temporal patterns and relationships across sensors. Real-world examples drawn from network monitoring and diverse time series datasets will illustrate how these approaches can be applied across different domains, highlighting their potential for anomaly detection, forecasting, and predictive maintenance. Finally, we will explore the promise of pretrained foundation models and zero-shot/few-shot learning, alongside practical limitations related to generalization, evaluation, false alarms, and computational cost. The emphasis is on concepts, real-world applications, and informed model selection rather than mathematical derivations or coding .

Learning outcomes:

• Distinguish forecasting, anomaly detection, and predictive maintenance, and recognize suitable industrial use cases.
• Understand the main challenges of real-world univariate and multivariate time series data and the strengths of classical and deep-learning approaches.
• Explain how generative AI and time series foundation models support time series modeling, including zero-shot and few-shot scenarios.
• Identify practical considerations for selecting and evaluating models, including data quality, false alarms, generalization, and computational cost.

Target audience

Industry practitioners, engineers, researchers, data scientists, and decision-makers interested in using AI to extract insights from time series data.

This webinar is open and free of charge for all participants from academia, industry, and public administration from EU and/or EuroHPC JU member countries.

Entry level & prerequisites

Beginner to intermediate. No prior expertise in time series modeling or deep learning required; basic familiarity with data analysis is helpful.

Agenda

Speaker

Dr. Pedro Casas (AIT Austrian Institute of Technology)              
Dr. Pedro Casas is a Senior Scientist at the AIT Austrian Institute of Technology, specializing in Data Science and Artificial Intelligence. He holds a degree in Electrical Engineering (2005) and dual Ph.D. degrees in Computer Science and Electrical Engineering (2010). His research focuses on machine learning and deep learning for complex and dynamic data, with particular emphasis on anomaly detection, forecasting, graph-based learning, multivariate time series analysis, and generalizable AI models.
He leads and contributes to applied AI research in networking (AI4NETS) and industrial applications (AI4Industry), addressing challenges in network monitoring, cybersecurity, predictive maintenance, fault detection, and industrial analytics. His recent research explores advanced AI approaches, including Graph Neural Networks (GNNs), generative AI, Time Series Foundation Models, and agentic AI, with applications in phishing detection, time series forecasting, and anomaly detection. His work emphasizes robust, generalizable, and data-efficient AI solutions across diverse application domains.
Dr. Casas has published more than 240 peer-reviewed papers in leading international conferences and journals and has received 20 awards recognizing his research contributions and project achievements.

Location

This course is offered as LIVE ONLINE WEBINAR via Zoom.

Registration

Registration is required, the link to the registration form can be found on the top of this page in the menu on the left. Please register with your official institutional email address to prove your affiliation.

You will get the Zoom link in the automatic confirmation by email (subject starting with "[Indico] Registration"), please check your Spam/Junk folders). We will also send 1–2 reminders before the webinar.

Please do not hesitate to contact us at training@ai-at.eu if you have any questions.

Modification, withdrawal & no-show policy

Your registration is binding. Please only register for the course if you are really going to attend.

You can update your registration data or withdraw your registration anytime before the registration form has been closed via the link "Manage my registration" which you can find at the bottom of your automatic email confirmation (subject starting with "[Indico] Registration").

Alternatively, or after the registration form has been closed, please inform us about your cancelation or any change in your registration data (especially your mobile-phone number) via email (training@ai-at.eu).

No-show policy: If you do not cancel and do not show up at the course you will be blacklisted and excluded from future training events.

Organizers

This webinar is organized by AI Factory Austria AI:AT.

Acknowledgements

AI Factory Austria AI:AT has received funding from the European High-Performance Computing Joint Undertaking (JU) under grant agreement No 101253078. The JU receives support from the Horizon Europe Programm of the European Union and Austria (BMIMI / FFG).