Anticipating network failures and congestion in optical networks a data analytics approach using genetic algorithm optimization.

Congestion-level forecasting is essential for optical links to function properly with networking management schemes. Effective dynamical routing and wavelengths assigning (EDRWA) difficulties under the presumptions of perfect and non-ideal structural levels represent some of the foremost challenging optimization challenges faced in constructing 3rd-generation optical links. Such two issues are each NP-complete by design. They are challenging since the search space contains several nearby optima. Heuristics-dependent methods cannot effectively handle such challenges because of the irregular nature of searching space. The route and spectral assigning problems under various demanding scenarios are solved in this study using a genetic method optimization method for creating reliable optical cores. In contexts with stochastic unpredictable needs, this enhances network efficiency. The model outcomes show how successful genetic algorithms are at optimizing dedicated optical spectrum allocation in the face of uncertainty. [ABSTRACT FROM AUTHOR]