Description

The topic of the lecture is the quantitative analysis of various types of data occurring in the field of transport and logistics. The focus will be on problems of maritime logistics. The students should be enabled to evaluate the learned methods with regard to their usability in concrete company contexts and to know and derive requirements and potentials of an effective application; for example related to data mining approaches for controlling or forecasting approaches for the operational planning of companies.

Within the course "Machine Learning in Logistics", students are introduced to a selection of methods of machine learning in the first half of the term. The lecture is called "Fundamentals of Machine Learning" and is organized by the Institute for Software Systems. In the second half of the term, the lecture "Digitalization in Traffic and Logistics" applies those methods to logistics-specific research questions. The course is accompanied by the exercise "Machine Learning in Logistics". Here, Jupyter notebooks are used to equip students with the optimal tools for data exploration.

Lecture contents

• The project structure for machine learning in science and industry
• Use cases for machine learning in logistics and transportation
• ML-supported machine vision (including deep learning) in road traffic and logistics
• Time-related data and movement data in transportation
• Automated anomaly detection