Best Student Paper Award for Work on Predictive Maintenance using Deep Learning

Date of news: 22 February 2022

The paper "Grouping of Maintenance Actions with Deep Reinforcement Learning and Graph Convolutional Networks”, by SWS Master student David Kerkkamp and Zaharah A. Bukhsh, Yingqian Zhang, and Nils Jansen, has won the best student paper award at the International Conference on Agents and Artificial Intelligence (ICAART) 2022. The paper is based on David's Master thesis at Rolsch Assetmanagement within the PrimaVera project under the supervision of Nils Jansen. The main contribution is a novel method for optimal maintenance planning of networked assets based on deep reinforcement learning and graph neural networks.