Sometimes it takes an outsider to pinpoint a pain point in an industry that’s set in its ways. In this segment of “Fix This,” we look at Countalytics’ solution to a perennial industry problem: manual inventory management is time consuming.
Inventory management isn’t as easy as 1, 2, 3. This became clear to David Hailey after working as an auditor at Ernst & Young and Delta Air Lines. He partnered with Robert McFadden, who has experience managing IT processes, to come up with an easy-to-implement solution.
The result? Countalytics was co-founded in 2016 – Hailey is chief executive officer and McFadden is chief operating officer. According to Hailey, most airlines that count inventory do so manually, and only a couple are working on an automated system, which can be expensive to develop internally.
“We have found that airlines do not have full visibility into what consumption levels are by flight,” Hailey says. Countalytics aims to patch the hole in the process by using IBM’s Watson – specifically, its visual-recognition and machine-learning capabilities – to automate inventory management.
“We have found that airlines do not have full visibility into what consumption levels are by flight.” — David Hailey, Countalytics
Countalytics must first populate its server with an airline’s inventory – thousands of images of each item are taken either by a photographer or by recording a video of the item rotating on an axis to capture various angles. “The more pictures we take, the better the visual recognition is,” Hailey explains. “That’s machine learning: The more images, the higher the accuracy levels.”
Once an airline’s inventory is represented in the server, flight crew can upload a photo of a trolley tray and the system returns an inventory count in less than 10 seconds that is over 90 percent accurate, and can also inform forecasting. For example: “If data from the last six months of inventory consumption on a specific flight shows that people didn’t want Coke Zero, maybe they can be removed, thereby reducing inventory costs and potentially even weight,” Hailey suggests.
Based on information gathered from flight attendant focus groups, the solution could offer cost savings from 10 to 35 percent by addressing consumption that’s not accounted for. “Our platform is up and running and ready to be implemented,” Hailey says. “We are currently talking to three carriers about rolling out a beta version.” But who’s counting?
“Fix This” was originally published in the 9.1 February/March issue of APEX Experience magazine.