Orateur
Description
Flexibility is a critical attribute in solving stochastic programming problems, where decision-making must account for uncertainty and adapt to a wide range of potential future scenarios. Stochastic programming is a practical approach in which decisions are determined prior to the realization of uncertain variables, with subsequent adjustments made through recourse mechanisms once the uncertainties are revealed. Traditional solving methods rely on scenario-based approximations, where incorporating additional scenarios may improve accuracy but simultaneously escalates computational complexity, often making large-scale problems computationally intractable.
This study aims to address these challenges by using neural networks as surrogate models to estimate solution quality across diverse scenarios, thereby reducing the computational burden when integrated into the optimization framework. Empirical evaluations have been conducted on single-source capacitated Facility Location Problems, a stochastic variation of the multi-path Traveling Salesman Problem. Preliminary results demonstrate the effectiveness of this approach taking advantage of the generalization capability of the neural network model.
