Description:

This course introduces data analytics and selected artificial intelligence methods with a strong focus on applications in supply chain management, logistics, transportation, production, and mobility systems. Students learn how to design, implement, and evaluate data-driven models for problems such as demand forecasting, inventory planning, travel-time and delay prediction, routing and fleet management, anomaly detection in production, and predictive maintenance. The course combines methodological content (supervised and unsupervised learning, time-series methods, evaluation, and links to optimization models) with hands-on work on realistic datasets using Python. Emphasis is placed on the full analytics pipeline: problem formulation, data understanding and preparation, model building and validation, and translation of model outputs into managerial decisions.

Learning Targets:

After successful completion of the course, students are able to:

• Describe and distinguish fundamental artificial intelligence methods and their assumptions;
• Formulate typical problems from supply chain management (e.g., demand forecasting, inventory planning, anomaly detection) and mobility (e.g., travel-time prediction) as machine learning tasks and develop appropriate solution approaches;
• Implement core artificial intelligence techniques using Python and relevant libraries;
• Evaluate and compare models using appropriate metrics and validation methodologies;
• Interpret model results and relate them to classical operations research problems.

Outline:

1. Artificial intelligence and applications in supply chain management and mobility
2. Selected supervised learning methods
3. Selected unsupervised learning methods
4. Deep learning
5. Reinforcement learning
6. Interpretability

Literature:

Methods of Assessment:

Exam, Duration: 60 Minuten. Bonus point (quizzes).