Fulfillment Optimization Problem Instances

该数据集包含了履行优化问题的实例，用于展示Picard迭代算法在供应链强化学习中的政策模拟效果。此外，数据集还包括了一个奖励随机变量的生成模型，以及初始库存和容量向量，这些足以处理80%的订单。数据集的规模属于中等偏大，包含30个节点（J = 30），1个产品（I = 1），以及3个订单时段（T = 3）。其任务是进行政策模拟与优化。

This dataset contains instances of order fulfillment optimization problems, which are used to demonstrate the policy simulation performance of the Picard Iteration algorithm in supply chain reinforcement learning. Furthermore, the dataset includes a generative model for the reward random variable, as well as initial inventory and capacity vectors, which are sufficient to handle 80% of orders. The dataset is of medium-to-large scale, comprising 30 nodes (J = 30), 1 product (I = 1), and 3 order periods (T = 3). Its task is policy simulation and optimization.