(Bloomberg) – Industrial production is one of the dirtiest corners of the corporate world. A startup of ex-Google engineers thinks it can clean it up with artificial intelligence.
Phaidra, a Seattle-based company, sells AI software to automate building controls for power plants and other industrial giants. It relies on the same patch as their old company, DeepMind, Google’s research lab. For several years, DeepMind has let its AI system manage temperature checks inside Google data centers, ultimately shaving off huge chunks of the company’s electricity bill.
Phaidra’s algorithms are designed to select the most efficient temperature for unique facilities, such as a steel mill or vaccine manufacturer, and identify when equipment begins to lag in performance. Once in place, Phaidra’s system can reduce a plant’s energy consumption by up to 30% and save considerable amounts of capital, according to the startup. “It can immediately make those businesses more profitable,” says Jeremy Brewer, managing partner at Starshot Capital, an investor.
Jim Gao, CEO of Phaidra, calls manufacturing an industry overlooked by Silicon Valley, but ripe for the kind of advanced machine learning being cooked up in places like Google. “They’ve been collecting data for so long, but they’re not using it,” he says of his new client base.
Indeed, the industrial sector, which accounts for about a quarter of all US greenhouse gas emissions and continues to grow, is beginning to embrace cutting-edge data science. A report by IOT Analytics projects that industrial AI revenue will reach $72.5 billion by 2025, up from $11 billion in 2018.
Yet most of this usage represents basic tasks like scanning data or creating online dashboards, not tools where algorithms run entire control systems without people changing dials, like this that Phaidra presents. Few manufacturers have the ability to attempt this or the budgets and engineering prowess to maintain such a system. “It’s very little,” says Jon Van Wyck, director general of BCG.
Phaidra, which was established in 2019, says it has several Fortune 100 industrial clients in areas as broad as pharmaceutical development to paper mills. He declines to name specific customers or financial figures. The startup recently raised $25 million in funding from Starshot and Character, investment firms created by other Google alumni. He also recruited Robert Locke, a 13-year veteran of industrial supplier Johnson Controls, as chairman.
Gao previously worked in the energy team at DeepMind alongside its technical co-founder, Vedavyas Panneershelvam. Engineers are among the few with extensive experience in reinforcement learning, a branch of AI where algorithms are designed to continually improve. The most famous version of DeepMind is AlphaGo, its system for whupping the popular challenging board game, Go.
While AlphaGo has been optimized to win the Go, Katie Hoffman, another Phaidra co-founder, describes her system as one optimized to reduce the kilowatt hours of the factories it plugs into. Hoffman comes from the industrial sector – most recently as a director at equipment manufacturer Ingersoll Rand Inc. – and says many of Phaidra’s customers work in “critical” areas, with very specific requirements on how their factories operate. must be cooled and operated. They also rely on outdated, hand-coded software.
“They’re using what’s been around since the 1950s,” she says. “These industrial systems are incredibly difficult to operate on a good day. »
DeepMind’s building control algorithms continued to change dials inside Google’s data centers. And Google has started offering a similar service to its cloud customers. But the founders of Phaidra say their approach is tailored to the particularities of the industrial sector, which is light years away from the sophistication of a Google building.
Gao did not share his company’s pricing, but says customers pay Phaidra less than the energy savings they get from its service.