Data-driven methods are transforming modern engineering, enabling smarter product design, more efficient production systems, and intelligent robotics. To prepare students for these challenges, this course emphasizes the applied, technical aspects of data-driven engineering.

Building on prior courses introducing data science and machine learning (see prerequisites), core content includes technical aspects of industrial data sources (e.g., database technologies, REST APIs), messaging technologies (e.g., MQTT), version control (e.g., git with GitLab), and server or cluster-based computations for scalable data processing (e.g., HPC at KIT). Students practice through guided exercises and work in small groups on applied machine learning case studies using real-world engineering data from our lab. The module also covers methods for model explainability and interpretability to support transparent decision-making in engineering applications.

The course is designed as an on-site project with regular in-person collaboration and close supervision, offering students continuous, structured, and individualized feedback on their code and technical implementation.

Prerequisite

“Data Science Fundamentals in Engineering”,  “Grundlagen der Künstlichen Intelligenz ” or a comparable course must be passed.

Learning Objectives

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

• Apply advanced data science and machine learning methods to engineering data sets, including data from industrial sources and streaming technologies.
• Design and implement data analysis and machine learning workflows using collaborative tools, version control, and cluster-based computing.
• Analyze, interpret, and evaluate the performance and limitations of machine learning models in applied engineering scenarios.

Additional Information

The number of participants is limited. Information on the application process is available in ILIAS.

Organizational Information

See further details on course organization at: lehre.imi.kit.edu and in ILIAS.

Workload

120 hours

See ILIAS for time and location