Deadline-constrained multi-commodity flow routing and scheduling optimization with consideration of edge lengths and capacities

• Published in: Computers & industrial engineering Vol. 192; p. 110193
• Main Authors: Wang, Yu, Kang, Rui, Guo, Linhan, Zhang, Chao, Deng, Jianhui, Liu, Pengpeng, Wen, Meilin, Le, Tian
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.06.2024
• Subjects: Deadline; Edge lengths; Multi-commodity flow; Optimization; Priority; Time-varying traffic; Multi-commodity flow; Edge lengths; Time-varying traffic; Priority; Optimization; Deadline
• ISSN: 0360-8352; 1879-0550
• DOI: 10.1016/j.cie.2024.110193

With the emergence of satellite communications, meeting deadline requirements for data flow has become more challenging due to the trade-off between route lengths and capacities, as traversal delay becomes more prominent in satellite networks. To address the transmission optimization problems of multi-commodity flow with deadlines in satellite-based networks like networks considering edge lengths and capacities, we propose a new traffic model of flow conservation with the delay of edge lengths and then construct the deadline-constrained multi-commodity flow routing and scheduling optimization based on the proposed traffic model. Besides, for priority scheduling of multi-commodity flow, admission control is studied to schedule the commodities according to the priority level based on the proposed deadline-constrained flow optimization model. To solve the proposed models, we utilize the time-slicing approach and decomposition methods. Flow optimizations and admission control are discretized into linear programming and mixed 0–1 linear programming, respectively, which are suitable for Benders and Dantzig–Wolfe decomposition. We provide numerical examples using the Iridium satellite network to demonstrate the effectiveness of our routing and scheduling methods compared to those without consideration of edge delays. •New flow conservation to describe time-varying traffic with traversal delay on edges.•Multi-commodity flow optimization subject to deadlines and flow conservation with edge delay.•Scheduling of dynamically incoming multi-commodity flow.•Admission control to maximize the total flow volume considering priorities.