In this article series we explore optimization of a "picking warehouse" – that is, a warehouse where we pick items off the shelves to fulfil a customer order. As a result, we improve the efficiency of our warehouse by a factor of three.

Specifically, we model the efficiency of:

• A variety of shelving layout designs.

• Positioning items on shelves according to order probability.

• Combining orders to achieve economies of scale.

Along the way, we assess many alternative warehouse designs, which requires solving more than a million shortest path problems (using Dijkstra's algorithm via the networkx library) and almost a million Travelling Salesman Problems (using the constraint solver in OR-Tools).

Unlike most of our blog articles, our focus isn't on the details of the model code – though the code and data are available to download. Instead, we focus on describing the steps we took to model the situation.

This first article describes and tests the main modelling setup. The subsequent articles will address our modelling questions.

To download the model and understand the formulation of this problem continue reading on the Solver Max website. You will find there many other fantastic optimization materials, optimization models and a great collection of similar websites.