How a 3PL Eliminated Shrink With Drone-Powered Inventory Counting

A case study on how Gather AI’s autonomous drones are helping a retailer of sports memorabilia to keep near-perfect track of its extensive high-value inventory, and virtually eliminate the problem of shrink within the warehouse.

featured-image

Founded in 2015, Stadium Goods is a New York City-based retailer of aftermarket collectible sports footwear, apparel and other memorabilia, stored in a New Jersey warehouse on a consignment basis. Each item is unique, with a single pair of shoes valued at anywhere from $50 to $15,000 or more. To manage that substantial inventory, Stadium Goods relies on Barrett Distribution Centers, a provider of third-party logistics, e-commerce fulfillment and distribution services.

For the retailer, Barrett stocks around 500,000 pairs of sneakers, each individually labeled, in high-density racking. As such, it’s critically important that the company maintain accurate inventory — a misplaced item is the same as a lost one. Stadium Goods’ previous warehouse provider was struggling to keep proper tabs on its inventory, with 1.



5% to 2.5% shrink — an unacceptable margin of error, given the high price tag on so many of the items. And a disappointed end user is a customer who won’t be a source of repeat business.

Manual cycle counting was out of the question. To cover the entire warehouse, six people were required to move slowly through the aisles, scanning each item by hand. And with Stadium Goods’ inventory turning at least once per quarter, the cost of that effort could run as high as $300,000 per year — not to mention the price of mistakes, with Stadium Goods on the hook to the consignor for any lost product.

Barrett turned for help to Pittsburgh-based Gather AI, which sells autonomous drones that use artificial intelligence and machine learning to automate a variety of warehouse tasks, and track inventory in real time. The drones — which require no external lighting to operate — create a digital map and 3D model of the entire warehouse. Using AI and computer vision to read the barcodes, they identify the precise location of each SKU, and directly compare it to what’s in the warehouse management system (WMS).

The information is loaded onto a web-based dashboard, creating a pictorial representation of each item. Barrett can click on any aisle or inventory location to ensure that everything matches up. In addition, a list of exceptions allows for the instant identification of any irregularities.

The whole system is managed by an iPad Pro. Implementation of the system was easy, according to the companies — Gather AI says it typically takes as little as a week and a half to get a facility completely mapped by its drones, with workers fully trained on the technology within two to three weeks. At no point does introduction of the technology impact the customer’s deliverables or slow down operations, Gather AI claims.

The technology means that warehouse workers no longer need spend long, tedious hours doing manual inventory counts, substantially reducing the likelihood of misplaced products, risk of over-ordering and delayed shipments. Warehouse managers can easily identify and fix inventory exceptions, even creating a to-do list for their teams. The system also helps warehouses to cut down on costs tied to the need for automated lifts, allowing users to conduct more frequent cycle counts with less equipment, as well as insurance and maintenance expense.

Gather AI says the drones perform at a rate of up to 15 times faster than manual cycle counting. In the Barrett facility, they’re scanning around 700 individual “license plates” per hour, seven times faster than the previous manual rate. Scanning costs are significantly lower as well; from $1.

50 per scan for a person using a radio-frequency gun, to between 10 and 20 cents when done by the drones. The technology made possible the reallocation of those six human cycle counters, saving $250,000 on material handling equipment. Gather AI sets a standard of 98% license-plate reading accuracy for its drones, but the result for Barrett and Stadium Goods was even better — 99.

9% this year to date at the New Jersey warehouse. And shrink — a major headache for Stadium Goods under its previous warehouse provider, has shrunk to near-zero. Barrett says the high level of scanning accuracy has the additional advantage of boosting warehouse employee satisfaction, eliminating the frustrations experienced when products aren’t where they belong.

Stadium Goods, Barrett Distribution Centers, Gather AI, RELATED CONTENT RELATED VIDEOS Related Articles Related Directories.