Skip to content

Machine learning for predictive maintenance

5 credits

Who is this course for?​

The course introduces machine learning methods for equipment condition monitoring. It is intended for people who already have the basics of machine learning and artificial intelligence, with basic skills in machine learning programming (Python), and with interest in understanding how these technologies can be used to design data-driven equipment monitoring systems.

What will you learn from this course?

Students will learn about the basics of equipment monitoring, deviation detection, and methods to predict the remaining useful life of a piece of equipment. The course focus is on equipment that exists in fleets, i.e. there are several more units than one. Examples of such equipment are vehicles, windmills, and engines. The course will provide several examples of such equipment monitoring.

What is the format for this course?

Instruction type: Lectures are delivered via a video conference tool and in English. In addition to lecture slides, complementary material such as Jupyter notebooks and examples of scientific papers will be provided as the basis for exercises, project work, and seminars.

Frequency: The course is broken down into 4-6 lectures, including some guest lecturers by professionals in the industry who have hands-on experience with equipment monitoring and machine learning in real applications, and 4 exercise sessions (Jupyter notebook) presented weekly over two months.

Examination: The students will demonstrate their knowledge by giving a seminar on a scientific paper and presenting a practical project.


Course responsible:

Thorsteinn Rognvaldsson