A novel learning model with dynamic heterogeneous graph network for uncertain multimode resource-constrained project scheduling problem

This paper addresses a multi-scenario multi-mode resource-constrained project scheduling problem with the goal of minimizing both the makespan and cost of the project. In order to visualize the changing process of modes and priority relationships in a project, a dynamic activity-mode network graph i...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Computers & operations research Ročník 186; s. 107319
Hlavní autoři:	Wang, Zheng, Liu, Huiran, Fan, Xiaojun
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Elsevier Ltd 01.02.2026
Témata:	Dynamic heterogeneous graph neural network Multimode resource-constrained scheduling Reinforcement learning algorithm Scenario-based uncertainty Trade-off time and cost Dynamic heterogeneous graph neural network Multimode resource-constrained scheduling Trade-off time and cost Scenario-based uncertainty Reinforcement learning algorithm
ISSN:	0305-0548
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	This paper addresses a multi-scenario multi-mode resource-constrained project scheduling problem with the goal of minimizing both the makespan and cost of the project. In order to visualize the changing process of modes and priority relationships in a project, a dynamic activity-mode network graph is introduced. Based on this network, a deep reinforcement learning model based on dynamic heterogeneous graph neural network is designed, and 12 solving models are obtained by training this model using the proximal policy optimization algorithm. The convergence of the model was verified using benchmark instances from the Project Scheduling Problem Library. Meanwhile, based on the characteristics of the solved problem, 360 instances are generated by reproducing the algorithm for generating benchmark instances. The problems are addressed using these 12 solution models and 9 additional comparison algorithms. Furthermore, a sensitivity analysis is conducted regarding the configuration parameters of the problem. The results validate the optimal effectiveness, stability, and generalization ability of the proposed learning model. It also demonstrates that this model can be a robustly better solving model and scheduling scheme according to actual demands. •Proposes a dynamic heterogeneous graph network model for uncertain multimode RCPSP optimization.•Combines graph networks and reinforcement learning to enhance project scheduling efficiency.•Outperforms Q-learning and meta-heuristic algorithms in terms of optimization and stability.•Captures real-time project activity data for better resource allocation decisions.•Offers practical solutions for project managers in construction, software, and manufacturing industries.
ISSN:	0305-0548
DOI:	10.1016/j.cor.2025.107319