Inventory Optimization: How to Handle Slow Movers
Do you know how many boxes of raspberry gourmet coffee Kroger should keep at their store in rural Georgia? How would that computation be different from the logic needed to calculate the number of boxes of Corn Flakes at the same store? Raspberry gourmet coffee is likely what is known as a slow mover. You may be wondering why the number of raspberry gourmet coffee boxes Kroger has in their inventory at their store in rural Georgia even matters, but it does and here’s why: Inventory management is a critical part of every company’s supply chain and is a large part of the operating costs in all industries.
Handling inventory of fast movers, items that are sold quickly and in large amounts, is very well understood. However, many open questions still arise when it comes to the other part of inventory: Slow movers. Think back to the raspberry gourmet coffee example. Slow movers are items that have very few customer demands, but need to be stocked nonetheless for a variety of reasons. They can make up a large proportion of a company’s assortment. Even though only a few units of every slow-moving item are stored, collectively they can take up a lot of inventory space. This is a pressing concern and leads to one of the most basic inventory management questions: How many to stock and when to order more?
The rise of the inventory management tools for fast movers came from the understanding of the underlying statistical models. Most fast movers can easily be modelled using a Gaussian distribution. For example, simple, average and standard deviation measures are sufficient to explain their behavior. Statistical models are more efficient when built on large datasets, and fast movers are a perfect fit since they are defined by their high demand history. Experienced inventory planners will tell you that you have to consider more than the sale numbers to determine the future inventory levels. JDA products handle into their planning process other factors such as weather, sports events and holidays. This allows companies to detect sudden and seasonal demands in advance.
From a statistical point of view, slow moving items are hard to forecast since low number of demands leads to a poor understanding of the demand patterns. At JDA Labs, we are developing a system to handle those items without using a forecast or standards statistical distributions. To circumvent the lack of statistical distributions, we use the historical demand directly. Each item history is handled individually to obtain a tailored statistical distribution.
This system also handles the bigger picture: Inventory managers want to satisfy target of performance and accuracy. This is why we support groups service levels. Service levels are a common performance measure for inventories. It is very difficult to ensure a service level for an individual item when dealing with the high uncertainty of slow moving items. Therefore, missing a service level for one item among a large inventory may not be a big issue, but it is important to make sure that on average the service level among a group of items match the target. Hence, our product finds the optimal inventory policy matching the target service level and minimizing the operating costs.
How JDA Labs Can Help
With the ongoing support of various co-innovating enterprises, JDA Labs is continuously refining the system and learning the relevant use cases our clients need. JDA Labs collaborates with other product development teams to transfer this knowledge inside the existing JDA products. Handling issues that supply chain leaders face every day with new innovative products is another way that JDA Labs delivers more value to the customers. You can learn more about Inventory Optimization and how JDA helps companies profitably balance speed and responsiveness with real-world inventory costs and risk exposure here.
Comments are closed.