Jonathan Hoss

Jonathan Hoss

PhD Candidate @ TH Rosenheim

Reinforcement Learning for Industrial Production Scheduling

I research scalable learning-based methods for scheduling and decision-making under real-world industrial constraints.

GitHub LinkedIn

About me

I hold a Master's degree in Industrial Engineering from Rosenheim University of Applied Sciences, graduating with the highest distinction (grade 1.0). With international experience in Malaysia and Finland, I bring a global perspective to my work. Currently, I am a PhD candidate at Rosenheim Technical University, specializing in reinforcement learning for production scheduling and exploring the integration of digitalisation, AI, and cloud technologies to advance innovative industrial solutions.

Outside of academia and research, I enjoy exploring the outdoors—especially paragliding in the mountains, where I also conduct tandem flights in cooperation with Tandemfliegen Chiemgau, sharing the experience with others.

Jonathan Hoss tandem paragliding over Kössen with his girlfriend
Tandem paragliding over Kössen.

Publications & Work

  • Best Paper Award Nominee (Top 3 out of 533 submissions)
    Hoss, J.; Link, M.; Klarmann, N. Scalable Production Scheduling: Linear Complexity via Unified Homogeneous Graphs. Presented at the 2026 IEEE 22nd International Conference on Automation Science and Engineering (CASE 2026), August 17-21, 2026, Shenyang, China. Proceedings forthcoming. preprint available on arXiv
  • Link, M.; Hoss, J.; Klarmann, N. An Analysis of the Coordination Gap between Joint and Modular Learning for Job Shop Scheduling with Transportation Resources. Presented at the 2026 IEEE 22nd International Conference on Automation Science and Engineering (CASE 2026), August 17-21, 2026, Shenyang, China. Proceedings forthcoming. preprint available on arXiv
  • Toro Diz, M.; Hoss, J.; Klarmann, N. Measurement-Calibrated Multi-Camera Fusion for Vision-Based Indoor Localization. Presented at the 2026 IEEE 22nd International Conference on Automation Science and Engineering (CASE 2026), August 17-21, 2026, Shenyang, China. Proceedings forthcoming. preprint available on arXiv
  • Hoss, J.; Klarmann, N. Bridging the Sim-to-Real Gap in Reinforcement Learning-Based Industrial Dispatching through Execution Semantics. Presented at the IEEE International Conference on Industrial Informatics (INDIN 2026), 26-29 July 2026. Proceedings forthcoming. preprint available on arXiv
  • Hoss, J.; Schelling, F.; Klarmann, N. A Production Scheduling Framework for Reinforcement Learning Under Real-World Constraints. 2025 IEEE 21st International Conference on Automation Science and Engineering (CASE 2025), 17-21 August 2025, Los Angeles, CA, USA, pp. 1736-1743. https://doi.org/10.1109/CASE58245.2025.11163982
Jonathan Hoss presenting at IEEE CASE 2026 in Shenyang, China
Presenting "Scalable Production Scheduling" at IEEE CASE 2026 in Shenyang, China.

What Matters to Me

Since 2020, I've been volunteering as Web & IT Administrator for Conambiki e.V., a charity dedicated to supporting education in Namibia. Together with an amazing team, we've raised over €400,000 to empower young learners and foster community development. It's meaningful work that connects technology with real-world impact, and I'm proud to contribute to making education more accessible.