Edge computing helps fuel Taco Bell’s digital business

At Taco Bell, a ubiquitous computer server collects data from physical and digital orders and customer loyalty tags, as well as kitchen operations, and uses custom algorithms to tell employees in the fryer when to dip. Mr. Pariser said the potatoes are hot when a delivery driver comes to pick up an order for nacho fries.

“We took it [on] “Our most critical workloads are on order and menu data,” Mr. Pariser said.

Linda Stadtmuller, senior vice president of the research, information and communications technology practice at market research firm Frost & Sullivan, said the edge is not a location, but a computer model.

Key aspects are sensors, connectivity, analytics and responsiveness, Ms. Stadtmuller said at the same event.

The goal is to improve the performance of applications where data is generated, such as at the local Taco Bell, and by applying it at lightning speed. Energy companies, retailers and industrial manufacturers are using edge computing to take advantage of faster internet speeds, 5G networks and the growing number of connected devices. General Electric Co.

and Siemens AG

For example, they are using edge computing to optimize factory machines in real time.

“The John Deere tractor is probably the edge when it comes to sensors that control parts. It can be an edge when your phone collects data or detects and plugs into a kiosk,” Ms. Stadtmuller said.

Taco Bell has spent about five years deploying edge computing capabilities, Mr. Pariser said. Each location has duplicate devices that serve as backups in case of outages.

Beyond walking into a storefront or using a drive-thru, Taco Bell customers can order their Mexican pizza and Chalupa Supers through the company’s website, mobile app, food delivery services and, in some geographies, by text message. In January, the chain rolled out a subscription service nationwide. For $10, customers can purchase a Taco Lover’s Pass to get tacos a day for 30 days. Menus are customized by location.

When a repeat customer places an order, the local restaurant system fetches relevant information from the cloud. Testing repeat purchases, typical order volume and unique items can help determine whether to offer custom deals, Mr. Pariser said.

In addition to processing and analyzing orders from multiple platforms, restaurants must simultaneously handle menu adjustments, food assembly and staff turnover, as well as information from multiple channels, Mr. Pariser said. “You do these events and you try to optimize them. We’re no different than manufacturing in some ways,” he said.

The team had to think about how to protect personal information about customers in the Taco Bell loyalty program. “For security reasons, we don’t want the data to be in the store,” he says, and it’s stored in the cloud. “You don’t put more information on the sidelines than you can handle.”

Taco Bell contracts with a third party for a platform to track overall edge computing activity, and Mr. Pariser said his own team has developed software to route data to various connected devices in kitchens.

With the Edge Computing Foundation in place, Taco Bell says it can experiment with connected robotic devices that can fry food, heat tortillas or pour drinks. “Now it’s a little more interesting,” he said.

Write Kim S. Nash at kim.nash@wsj.com