4-Day Course: Data, AI and Multimodal Traffic Management

Date	2-3 October 2025 (block 1) and 17-18 November 2025 (block 2)
Time	9:00-17:00 h
Location	Mondai House of AI, Molengraaffsingel 29, Delft
Course Leaders	Prof. Dr. Ir. Serge Hoogendoorn,  Dr. Ir. Sascha Hoogendoorn-Lanser, Prof. Dr. Ir. Hans van Lint, Dr. Ir. Marco Rinaldi, Dr. Yanan Xin
Course fee	€ 975 (lunches and coffee/tea included)
Registration	Register here…

This four-day course brings together PhD researchers and mobility professionals to explore how artificial intelligence and data can be used to manage complex, multimodal traffic systems. The course offers an intensive yet accessible introduction to the key functions of traffic management—such as traffic state estimation, short- and long-term prediction, anomaly detection, real-time control, and optimization—and how modern AI methods can support these tasks.

Through a mix of thematic lectures, hands-on data sessions, and collaborative group work, participants will gain both theoretical understanding and practical insight into the role of AI in mobility. Real-world use cases, contributed by professionals, provide the basis for group assignments in which PhDs and practitioners work together to solve traffic problems using AI-based approaches.

A distinctive element of the course is the integration of Knowledge Capsules: short, focused modules developed and presented by small teams of PhDs. These capsules are designed to introduce core AI topics—such as clustering, reinforcement learning, or explainable AI—in formats that are directly relevant to traffic applications. Capsules may take the form of mini-lectures, short video tutorials, annotated datasets, or reproducible code examples, and serve as bridges between academic research and operational practice.

Program

The program is split into two blocks of two days each:

Block 1 (Days 1–2) – Foundations

• Traffic management functions and real-world applications
• Introduction to relevant AI techniques (e.g., supervised/unsupervised learning, time series forecasting, reinforcement learning)
• Hands-on data session
• Knowledge Capsules presented (optional, PhD-led)
• Group formation and case selection

Interim Period (5 weeks)

• Mixed PhD–professional teams work on use case assignments
• Optional further development of Knowledge Capsules

Block 2 (Days 3–4) – Application & Reflection

• Group presentations and feedback
• Problem-solving roundtables and expert panels
• Optional advanced capsules (e.g., “XAI for policymakers”)
• Final synthesis and takeaways

Course material

• Selected readings and tutorials on AI methods in mobility
• Example datasets (e.g., loop detectors, FCD, PT data)
• Existing NM Magazine tutorials
• Materials and slides from lectures and capsule sessions
• Optional: annotated code and notebooks for exploration

Methodology

The course blends academic and applied learning through:

• Thematic and interactive lectures
• Group work on real-world traffic use cases
• Peer-to-peer learning via optional Knowledge Capsules
• Hands-on sessions with traffic data
• Expert and peer feedback on outputs

Prerequisites

• Background in transportation, traffic, or AI-related fields
• Basic familiarity with machine learning concepts and Python is recommended