It’s not uncommon today to see an empty store shelf when shopping. As a consumer, this can be a frustrating experience, especially if the item you need is out of stock. And from a supply chain perspective, an empty shelf is even more alarming, as it represents lost revenue – for both the retailer and the manufacturer. Limited shelf space and a rise in new product introductions have only compounded the problem, making it harder for retailers to keep shelves stocked.
Why has this been a difficult problem to solve?
Of course, one way to solve the problem is to hire more staff to take stock of inventory and replenish empty shelves more regularly. This, however, is a costly way to solve the problem. It is expensive to deploy additional resources to review inventory positions more often, and many retailers have struggled to find the perfect balance between investing in labor and realizing a return on that investment in the form of incremental revenues.
Unlocking the value of image recognition for shelf logistics
As store designs change over time, retailers and manufacturers look for ways to offer a wider assortment of products to drive higher revenues per square foot of space. And the sooner retailers can detect an empty shelf, the sooner they can take action to replenish the shelf, preventing customer dissatisfaction and lost sales.
Image recognition technology provides companies with a new way to track inventory by validating share-of-shelf compliance, identifying stock-out situations, and ensuring planogram compliance – without adding any additional staff. This is a critical point, as many retailers are already operating at thin margins. Any marginal improvements in out-of-stock situations go a long way in bolstering the profitability of a store.
New technology ushers in new opportunities
Image recognition technology has come a very long way in recent years. Your favorite photo website likely uses image recognition software to help identify people, and in some cases, favorite locations. It accomplishes this through supervised machine-learning techniques and deep-learning algorithms that can analyze and comprehend images, much the same way a human learns to recognize the same object from multiple points of view.
In the case of shelf logistics, machines can be trained to recognize images a specific type cell phone or gadget at the local computer retailer. This occurs via a camera that routinely captures images of a shelf, and algorithms that can “see” a shelf is empty and needs to be replenished. This monitoring can happen in real time, without the need for staff to take inventory counts at predefined intervals. Machine-learning algorithms in this space are constantly evolving and the pace of learning is accelerating every day.
Image recognition technology is available in the market today. Leading supply chain software companies are coupling image recognition techniques with shelf logistics and supply chain management principles to deliver solutions that enable superior customer service, while driving higher revenues and profitability.
Image recognition for shelf logistics is just one way machine learning can be leveraged to power your digital supply chain. Learn how it can also enable intelligent segmentation, and drive insights from unstructured and structured data.