Pepsi Benefits from Artificial Intelligence, Machine Learning

PURCHASE, N.Y. – Robots, AI and machine learning are no longer just a competitive advantage for PepsiCo, according to Forbes. The $65 billion food-and-beverage giant is using those tools to sell more products in 200-plus countries.

In Stockton, Calif., the company has created a way for time-strapped students to get snacks delivered conveniently. Several Snackbots, six-wheeled, mobile vending machines, travel around the University of the Pacific campus, delivering Hello Goodness products, a better-for-you line of snacks, and sparkling water. The self-driving robots, a partnership between Robby Technologies and Pepsi, deliver items that students order from the Snackbot app. There is no delivery charge. The bots can travel 20 miles on a single battery charge and navigate at night, in rain or up curbs, thanks to onboard headlights and all-wheel drive capabilities.

PepsiCo has used another robot to help fill open positions. Robot Vera, as the technology is named, has phoned and interviewed candidates for jobs in sales, as drivers and to fill factory vacancies in Russia when the company needed to hire 250 employees in two months. Developed by Russian startup Stafory, Vera can interview 1,500 candidates in nine hours, a job that would take a human nine weeks.

Advanced speech recognition software, along with tools from Amazon, Google, Microsoft and Russian technology company Yandex, allow Vera to screen candidates for jobs, such as fork-lift operators, factory workers and sales. Its software can scan resumes to determine if a candidate has the right experience, respond to yes and no answers, ask questions and follow-up with correspondence. It also can forward transcripts of a call to a human HR specialist for review. So far, the majority of potential job candidates has responded positively to the interviews. There was a bit more hesitation from human HR professionals.

Another technology platform named Ada provides “augmented intelligence” by using human’s insight in combination with algorithms to lead to rapid learning. Ada can pull together data from a variety of sources, allowing PepsiCo to better leverage the massive amounts of data it collects. Ultimately, Ada will be involved in all aspects of operations, including innovation, design, research and pricing. The company also will put data to use in product development and sales and marketing.

Pep Worx, a cloud-based data and analytics platform, helps the company advise retailers on the best items to stock, where to place them and what promotions to use. When launching Quaker Overnight Oats, Pepsi identified 24 million households from a dataset of 110 million U.S. households and targeted the shopping venues that those households would be more apt to use. The company used data to create promotions aimed at those consumers and credit the technology with driving 80% of the product’s sales growth in the first 12 months after launch.

Manufacturing plants of Frito-Lay, a PepsiCo subsidiary, benefit from machine learning to automate the quality check for the chip-processing system. One project uses lasers to determine the texture of chips being produced. After the laser hits the chips, algorithms process the sound that is made and determine the chip texture.

Not stopping there, PepsiCo plans to offer a global training course on advanced machine learning and computer vision for its internal research and development team to continue finding insights that will drive efficiencies in its manufacturing facilities.