EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

ARIMA Forecasting and Linear Programming with Simulated Annealing for Adaptive Network Traffic Scheduling

Junfeng Dan, Yanhao Fan, Guo Chen, Zicheng Meng

Journal of Computer Science and Artificial Intelligence · 29.05.2026 · 7 (2) · 5-10

DOI: 10.54097/c19srf59 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computer Science and Artificial Intelligence, online erschienen am 29.05.2026. Verfasst von Junfeng Dan und 3 weiteren Autor:innen. Die Arbeit nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Adaptive resource scheduling in distributed network systems requires accurate forecasting of incoming traffic, principled allocation of workload across processing paths, and disciplined selection of high-impact nodes for capacity expansion. This paper develops a four-stage framework that integrates ARIMA time-series forecasting, single-objective and multi-objective linear programming, simulated annealing optimization, and entropy-weighted TOPSIS evaluation to handle the full scheduling pipeline. Stationarity is verified through ADF testing before fitting an ARIMA model that produces 31-day-ahead forecasts of per-path traffic volume with calibrated residuals. The forecasts feed a linear programming formulation that minimizes total operational cost subject to per-node utilization caps of 100%, solved through simulated annealing with adaptive cooling schedule. The multi-objective extension introduces auxiliary capacity reallocation goals and identifies one additional path 3→1 in the recommended topology. Entropy-weighted TOPSIS evaluation across multiple importance criteria selects the top three nodes 10, 14, and 4 and the top three paths 14→8, 14→9, and 36→4 for capacity expansion. Sensitivity analysis under random perturbation injection confirms that the proposed scheduling policy retains feasible utilization profiles across all perturbation scales tested, demonstrating robust performance under stochastic traffic variation typical of operational network deployments.

Zitieren

APA

Dan, J., Fan, Y., Chen, G., & Meng, Z. (2026). ARIMA Forecasting and Linear Programming with Simulated Annealing for Adaptive Network Traffic Scheduling. Journal of Computer Science and Artificial Intelligence, 7(2), 5-10. https://doi.org/10.54097/c19srf59

Vancouver

Dan J, Fan Y, Chen G, Meng Z. ARIMA Forecasting and Linear Programming with Simulated Annealing for Adaptive Network Traffic Scheduling. Journal of Computer Science and Artificial Intelligence. 2026;7(2):5-10. doi:10.54097/c19srf59

BibTeX
@article{Dan2026,
  author = {Dan, Junfeng and Fan, Yanhao and Chen, Guo and Meng, Zicheng},
  title = {ARIMA Forecasting and Linear Programming with Simulated Annealing for Adaptive Network Traffic Scheduling},
  journal = {Journal of Computer Science and Artificial Intelligence},
  year = {2026},
  volume = {7},
  number = {2},
  pages = {5--10},
  doi = {10.54097/c19srf59}
}
Kostenlos starten →