The research field "Vehicle Systems and Technology Assessment" offers a master thesis, in which the spatial emissions of passenger road transport worldwide are to be modeled within the project "EmissionsLandKarte (ELK)". The aim of the thesis is to develop a machine learning based model to predict regional passenger transport demand in terms of origin-destination matrices as an input to a fine-grained global transport emission inventory. Explainable neural networks and graph neural network architectures are to be conceptualized, tested and validated for Europe and then tested on other world regions. The collection and evaluation of appropriate input data from open sources for the model will be a first task. 

Your tasks: 

• literature research on existing algorithms for predicting regional transport demand and training datasets available
• input feature data collection and aggregation to appropriate spatial regions
• development and testing of suitable machine learning algorithms with a focus on explainability
• validation on other world regions to check for transferability

DLR is certified as a family-friendly employer. We offer you flexible working hours and temporary home office. A women's network has been established in our institute. Are you committed, able to cooperate, conscientious and responsible? Then apply to us!