Resource Scheduling Algorithm for Edge Computing Networks Based on Multi-Objective Optimization
Applied Sciences · 09.10.2025 · 15 (19) · 10837
DOI: 10.3390/app151910837 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 09.10.2025. Verfasst von Wenrui Liu und 7 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the 2024 Sci-Tech Innovation Project of China Tower (Nanjing) Science and Technology Innovation Center, Jiangsu Key Development Planning Project, the National Key R&D Program of China.
Abstract (Originalsprache)
Edge computing networks represent an emerging technological paradigm that enhances real-time responsiveness for mobile devices by reallocating computational resources from central servers to the network’s edge. This shift enables more efficient computing services for mobile devices. However, deploying computing services on inappropriate edge nodes can result in imbalanced resource utilization within edge computing networks, ultimately compromising service efficiency. Consequently, effectively leveraging the resources of edge computing devices while minimizing the energy consumption of terminal devices has become a critical issue in resource scheduling for edge computing. To tackle these challenges, this paper proposes a resource scheduling algorithm for edge computing networks based on multi-objective optimization. This approach utilizes the entropy weight method to assess both dynamic and static metrics of edge computing nodes, integrating them into a unified computing power metric for each node. This integration facilitates a better alignment between computing power and service demands. By modeling the resource scheduling problem in edge computing networks as a multi-objective Markov decision process (MOMDP), this study employs multi-objective reinforcement learning (MORL) and the proximal policy optimization (PPO) algorithm to concurrently optimize task transmission latency and energy consumption in dynamic environments. Finally, simulation experiments demonstrate that the proposed algorithm outperforms state-of-the-art scheduling algorithms in terms of latency, energy consumption, and overall reward. Additionally, it achieves an optimal hypervolume and Pareto front, effectively balancing the trade-off between task transmission latency and energy consumption in multi-objective optimization scenarios.
Förderung
- the 2024 Sci-Tech Innovation Project of China Tower (Nanjing) Science and Technology Innovation Center
- Jiangsu Key Development Planning Project
- the National Key R&D Program of China
- Natural Science Foundation of Jiangsu Province (Higher Education Institutions)
Zitieren
APA
Liu, W., Zhu, J., Li, X., Fei, Y., Wang, H., Liu, S., Zheng, X., & Ji, Y. (2025). Resource Scheduling Algorithm for Edge Computing Networks Based on Multi-Objective Optimization. Applied Sciences, 15(19), 10837. https://doi.org/10.3390/app151910837
Vancouver
Liu W, Zhu J, Li X, Fei Y, Wang H, Liu S, et al. Resource Scheduling Algorithm for Edge Computing Networks Based on Multi-Objective Optimization. Applied Sciences. 2025;15(19):10837. doi:10.3390/app151910837
BibTeX
@article{Liu2025,
author = {Liu, Wenrui and Zhu, Jiale and Li, Xiangming and Fei, Yichao and Wang, Hai and Liu, Shangdong and Zheng, Xiaoyao and Ji, Yimu},
title = {Resource Scheduling Algorithm for Edge Computing Networks Based on Multi-Objective Optimization},
journal = {Applied Sciences},
year = {2025},
volume = {15},
number = {19},
pages = {10837},
doi = {10.3390/app151910837}
}
EUVIMED