On LinkedIn, I came across this short, one and a half minute, video on how Amazon arranges items in its prime warehouses. These are the warehouses where Amazon stocks items in local markets for their 2- and 1-hour deliveries. The warehouses store items randomly throughout the warehouse to minimize the time to pick. The video is worth a watch.
Here are three quick lessons from the video:
- Optimization technology (like linear and integer programming) can drive make your operations more efficient. The heart of this video discusses an optimal routing that is given to each warehouse picker to minimize their time. There are many such operations that can be improved with optimization.
- I bet the placement of items isn’t random. The video makes it sound like Amazon randomly places items in the warehouse. This could be the case (I have no inside information here), but it is more likely that they are using machine learning algorithms to make sure items are placed near other items that are likely to be picked at the same time. And, they may even be using some optimization to then disperse the items so that the expected picking times are minimized.
- Technology allows you to do new and creative things. The optimization and machine learning algorithms are interesting. But, it is also interesting that Amazon challenged two long-held assumptions to improve operations: First, like items no longer need to be next to each other. Second, you can store individual products throughout the warehouse and on the same shelf as completely different items. Before if you did this, you couldn’t find anything.