Evolutionary Algorithms for Fibers Upgrade Sequence Problem on MB-EONs

Elastic optical network (EON) technology proves to be a promising solution to meet the ever-increasing bandwidth demands of Internet traffic. The transition from single-band EON (SB-EON) to multi-band EON (MB-EON) is a complex, multi-stage process, leading to the coexistence of fibers with different transmission capacities (SB and MB) and the formation of a hybrid SB/MB-EON for an extended period. This paper addresses the fiber upgrade sequence problem (FUSP), aiming to find the optimal upgrade sequence with the minimum average weighted load ratio (AWLR) while ensuring uninterrupted service throughout the entire process. We propose two evolutionary algorithms, including a genetic algorithm (GA) and a simulated annealing (SA), to solve this problem. Our proposed algorithms outperform conventional heuristics. GA demonstrates superior performance compared to SA in small networks or those with heavy traffic.