How to Avoid the Inventory Paradox in Supply Chain Network Modeling

Hi everyone, my name is Jeff Zoroya. I’m a principal in the supply chain design practice at Chainalytics. Inventory is an effective supply chain tool but it can also be very costly. You can leverage inventory to make your supply chain more responsive to changes in demand, and less prone to product shortages. However, inventory ties up working capital and increases your storage requirements. So it’s important to get it just right when you are evaluating changes to your supply chain network.

00:32 - There’s just one catch: This optimization problem is easier said than done. Let’s talk about how to avoid the inventory paradox in supply chain network modeling. First, it’s important to understand the relationship between inventory, storage space, and your network configuration. Determining this relationship is important. Where you source products and how those products flow through your supply chain affects the amount of inventory you will need and the amount of storage space you’ll need to house that inventory.

In most network modeling applications 01:04 - this relationship between inventory levels and throughput can be modeled as a non-linear relationship. This technique supports the basic inventory principle that a distribution network with more nodes will require more inventory, than one with fewer nodes. Structuring the relationship in this way makes it easy for the network modeling application to optimize the cost trade-offs between inventory, storage space, transportation, and other elements contained in your model. However, this approach breaks down when you’re considering network scenarios that aren’t represented in your historical data. In fact, the more pronounced the changes you’d like to make to your network, the less you can depend on the accuracy of this approach. Which brings us to the paradox…

01:50 - Supply chain modelers and fellow math enthusiasts are probably familiar with the chicken and egg paradox of combined network and inventory optimization. Generally, you can only optimize one business problem at a time. You can either design the best network configuration based on projected inventory behavior or you can determine the right amount of inventory to hold given your network configuration, re-supply parameters, and desired service levels. But you can’t do both. At least not at the same time. Most practitioners tackle this problem by using some type of an iterative approach where the network is optimized, excluding inventory and storage requirements, and then the inventory and storage is optimized based on that network configuration. Although this solves the problematic circular logic, an iterative approach such as this can leave some money on the table because you’re not optimizing inventory, storage space, and the other elements of supply chain cost simultaneously.

A single model allows you to 02:54 - leverage the power of the optimization. We at Chainalytics have developed an approach where we use inventory optimization to determine the likely inventory levels based on suitable representative network configurations. Then we include these inventory inputs in the network optimization. This approach considers all of the cost components in a way that avoids the chicken and the egg paradox and enables you to simultaneously optimize inventory, storage space, and your broader supply chain footprint. If you’re interested in learning more about Chainalytics’ single model approach, please reach out.

03:33 - Otherwise, I hope you have found this information helpful and learned something new. .