构建动态库存优化系统：深入探讨

In e-commerce, optimising replenishments is a crucial inventory problem. This involves solving three sub-tasks: What articles should be in stock? When should they be replenished? Where should the inventory be optimally allocated in the network of warehouses?

在电子商务中，优化补货是一个关键的库存问题。这涉及解决三个子任务：哪些商品应该有库存？何时应该补货？库存应该在仓库网络中如何最优分配？

Moreover, most e-commerce supply chains involve complex environments:

此外，大多数电子商务供应链涉及复杂的环境：

• Vast catalogue: up to millions of articles
• 庞大的目录：多达数百万个商品
• Multi-echelon network: dozens of warehouses spread across several countries
• 多层网络：分布在多个国家的数十个仓库
• Diverse and rotating catalogue: seasonal goods rotating on pre-defined and specific windows of sale
• 多样化和轮换的目录：季节性商品在预定义和特定的销售窗口中轮换
• High uncertainty****on key decision factors: Fluctuating demand patterns, and fluctuating shipment or supplier lead times.
• 关键决策因素的高不确定性：需求模式波动，以及发货或供应商交货时间波动。

At ZEOS, we recognise that our partners share these challenges. To empower them, we're developing AI-driven replenishment recommendations.

在 ZEOS，我们认识到我们的合作伙伴面临这些挑战。为了赋能他们，我们正在开发基于 AI 的补货建议。

The scale and complexity that this inventory problem brings makes it a unique combined Applied Science and MLE problem to solve. How can a system that continuously updates decisions consider these constantly changing and uncertain factors? The answer lies in building a dynamic inventory optimisation system.

这个库存问题所带来的规模和复杂性使其成为一个独特的应用科学和MLE问题。一个持续更新决策的系统如何考虑这些不断变化和不确定的因素？答案在于构建一个动态库存优化系统。

The article will cover:

本文将涵盖：

• Brief overview of the inventory optimisation framework
• 库存优化框架的简要概述
• Deep dive into how we scale demand forecasting and accelerate research in our demand forecasting pipelines
• 深入探讨我们如何扩展需求预测并加速需求预测管道中的研究
• Deep dive into how we run optimisation at scale in our policy optimisation pipelines
• 深入探讨我们如何在政策优化管道中进行大规模优化

Optimisation framework

优化框架

We frame replenishment decisions as a cost-optimisation exercise, with the end goal of minimising inventory costs:

我们将补货决策框架视为一个成本优化的过程，最终目标是最小化库存成本：

Min Costs(θ)=Cstorage(θ)+Clost sales(θ)+Coverstock(θ)+Coperations(θ)+Cinbound(θ)Mi...