AI Models Research and Development 

RWTH 
Sylwia Olbrych  Alexander Nasuta  Johanna Lauwigi 
Type: Consultancy / Online Tool / Software Tool
TRL: 3-6

WZL-IQS of RWTH Aachen University specializes in developing customer-tailored AI models, leveraging deep expertise in data-driven methods such as data analysis, Machine Learning, Reinforcement Learning, and Deep Learning—particularly in manufacturing. This expertise is one of our greatest assets, enabling us to deliver significant value to our customers. Additionally, we provide training and consultancy services to students, industry partners, and the broader community, fostering collaboration and innovation within the AI and Data Science communities.

IPR / Licence

MIT License

Contact Person

Sylwia Olbrych; Alexander Nasuta; Johanna Lauwigi

Information

The FAIRWork AI Service Catalogue can be found at:

• https://github.com/Alexander-Nasuta/FAIRWork-AI-Service-Catalog?tab=readme-ov-file

More information is provided in

• our D3.3, to be precise
  • chapter 3.6.2: Research on Existing AI Applications: ML Catalogue
  • chapter 3.6.3: Guidelines and Recommendations for AI Developers
  • chapter 3.6.4: AI-Based Optimizing Solutions for Industry
• the list of AI Services in D4.3:
  • Chapter 3.5.3: Resource Allocation using Neural Networks
  • Chapter 3.5.4: Resource Allocation using Linear Sum Assignment Solver
  • Chapter 3.5.5: Production Planning Service with a Hybrid Approach

Or you can have a look at our papers:

• Nasuta, A., Kemmerling, M., Lütticke, D., Schmitt, R.H. (2024). Reward Shaping for Job Shop Scheduling. In: Nicosia, G., Ojha, V., La Malfa, E., La Malfa, G., Pardalos, P.M., Umeton, R. (eds) Machine Learning, Optimization, and Data Science. LOD 2023. Lecture Notes in Computer Science, vol 14505. Springer, Cham. https://doi.org/10.1007/978-3-031-53969-5_16
• Olbrych, S., Nasuta, A., Kemmerling, M., Abdelrazeq, A., Schmitt, R. (2024). From Simple to Sophisticated: Investigating the Spectrum of Decision Support Complexity with AI Integration in Manufacturing. In: Lucas Paletta (eds) Cognitive Computing and Internet of Things. AHFE (2024) International Conference. AHFE Open Access, vol 124. AHFE International, USA. https://doi.org/10.54941/ahfe1004711
• Nasuta, A., Kemmerling, M., Zhou, H., Abdelrazeq, A., Schmitt, R. H. (2025). Curiosity Driven Reinforcement Learning for Job Shop Scheduling. Proceedings of the 17th International Conference on Agents and Artificial Intelligence, SCITEPRESS - Science and Technology Publications, 216–227. Porto, Portugal. https://doi.org/10.5220/0013143800003890

Use

For access to a live demo of our services, reach out to our designated contact person—we’re excited to help you get started!

Learn more about our Production Planning Service with a Hybrid Approach in the video below.

This service addresses the use-case of assisting with decisions about production planning, where recommendations for resource allocation are made. Based on state-of-the-art ML, this service provides a suggestion for a two-week planning horizon in two steps. First, it solves a job-shop problem by developing a production schedule that includes which parts are to be developed on which machine at which time and in which order. Second, it assigns workers to the tasks. This combination can offer a full production plan based on an order list and shift plan. It combines Constraint Programming for job shop scheduling and Monte Carlo Tree Search as a Reinforcement Learning approach for worker allocation. This hybrid methodology ensures an optimised production schedule while efficiently assigning workers to tasks, balancing machine utilisation and workforce constraints.

Explore our reports and GitHub repository for a deeper dive into the technical details.