English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The Flexible Job Shop Scheduling Problem (FJSSP) has been extensively studiedin the literature, and multiple approaches have been proposed within theheuristic, exact, and metaheuristic methods. However, the industry's demand tobe able to respond in real-time to disruptive events has generated thenecessity to be able to generate new schedules within a few seconds. Amongthese methods, under this constraint, only dispatching rules (DRs) are capableof generating schedules, even though their quality can be improved. To improvethe results, recent methods have been proposed for modeling the FJSSP as aMarkov Decision Process (MDP) and employing reinforcement learning to create apolicy that generates an optimal solution assigning operations to machines.Nonetheless, there is still room for improvement, particularly in the largerFJSSP instances which are common in real-world scenarios. Therefore, theobjective of this paper is to propose a method capable of robustly solvinglarge instances of the FJSSP. To achieve this, we propose a novel way ofmodeling the FJSSP as an MDP using graph neural networks. We also present twomethods to make inference more robust: generating a diverse set of schedulingpolicies that can be parallelized and limiting them using DRs. We have testedour approach on synthetically generated instances and various public benchmarksand found that our approach outperforms dispatching rules and achieves betterresults than three other recent deep reinforcement learning methods on largerFJSSP instances.

Solving large flexible job shop scheduling instances by generating a diverse set of scheduling policies with deep reinforcement learning