Large-Scale Inventory Optimization: A Recurrent Neural Networks–Inspired Simulation Approach
针对包含数千种成品和数十万种原材料的大规模生产网络，提出一种受循环神经网络启发的仿真方法，比现有方法快数千倍，能在合理时间内解决复杂库存优化问题。
Many large-scale production networks include thousands of types of final products and tens to hundreds of thousands of types of raw materials and intermediate products. These networks face complicated inventory management decisions, which are often too complicated for inventory models and too large for simulation models. In this paper, by combining efficient computational tools of recurrent neural networks (RNNs) and the structural information of production networks, we propose an RNN-inspired simulation approach that may be thousands of times faster than the existing simulation approach and is capable of solving large-scale inventory optimization problems in a reasonable amount of time. History: Accepted by Bruno Tuffin, Area Editor for Simulation. Funding: This work was supported by the National Natural Science Foundation of China [Grant 72091211].