University of Twente Student Theses
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.
Applying Deep Reinforcement Learning to solve the Flexible Job Shop Scheduling Problem
Slutter, C.P.E. (2025) Applying Deep Reinforcement Learning to solve the Flexible Job Shop Scheduling Problem.
|
PDF
5MB |
| Abstract: | Limis Planner is a rule-based scheduling approach for discrete manufacturing companies. The aim of this research is to explore the potential of Artificial Intelligence (AI) in machine scheduling and to compare its performance with Limis Planner and the Earliest Due Date dispatching rule. The machine scheduling problem that is central to this research is the (Flexible) Job Shop scheduling Problem ((F)JSP). The AI scheduling approach that is applied to solve this problem is adapted from a cutting-edge end-to-end policy-based Deep Reinforcement Learning (DRL) model from literature to include a minimum total weighted tardiness objective and manual job priorities. Besides, an alternative model is developed that incorporates a flexible machine availability extension by introducing new scheduling logic. The comparison between all scheduling approaches shows that the DRL approach is able to match the tardiness performance of the rule-based approaches in some of the less due date tight instances, although it is not as good in prioritising higher priority jobs. In cases of high due date tightness, the DRL approaches are making different trade-offs between focusing on minimising tardiness and prioritising higher priority jobs. The most suitable scheduling approach therefore depends on the preference of the end user of the approach. |
| Item Type: | Essay (Master) |
| Faculty: | BMS: Behavioural, Management and Social Sciences |
| Subject: | 31 mathematics, 54 computer science |
| Programme: | Industrial Engineering and Management MSc (60029) |
| Link to this item: | https://purl.utwente.nl/essays/108022 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page