Paul Stein, director of research and technology at Rolls Royce, gave a brief glimpse of the manufacturing giant’s efforts with Industrie 4.0 and AI today.
Speaking at the FT Innovate conference in London, Stein said that the challenge for the company is to innovate “but not too fast”, with more focus on “incremental, baby steps rather than major changes.”
Rolls Royce – innovating on services
Stein, who was part of a stellar panel including SwiftKey founder and CEO Ben Medlock and Open Data Institute chairman and co-founder Sir Nigel Shadbolt, admitted in his opening remarks that the popular ‘fail fast’ method of innovation doesn’t necessarily work at Rolls Royce. After all, there is little room for error at 20,000 thousand feet.
“What the customer wants is evolution, not revolution. The markets we sell into demand that tomorrow’s products are a bit better than today.”
Subsequently, he says that the modern airplane is not so different from the 1950s, adding that – with each aircraft typically in operation for 14 hours a day – adding new technologies into the airplane should be approached “with extreme caution.”
He says that 95 percent of Rolls Royce’s innovation strategy is about driving change inside-out, largely relying on existing staff to come up with solutions to fixing a specific problem. This, he says, requires “self-discipline” and a focus on utilizing technology to fix a particular problem.
Yet in services, it is a different approach entirely, with Stein believing that machine learning algorithms can – and should – be used to learn how an engine should work, and if it deviates from the norm.
“The application of innovation to services can be extremely rapid,” said Stein.
Industrie 4.0 live in the UK
Rolls Royce is of course no stranger to technological innovation. The firm is working with Microsoft and utilizing its Azure IoT cloud suite to improve aircraft efficiency and availability while lowering engine maintenance costs.
As reported on IoB earlier this year, the aircraft engine manufacturer uses the Azure IoT Suite to collect and aggregate data from disparate, geographically distributed sources and the Cortana Intelligence Suite to uncover new data insights. Data sets like engine health data, air traffic control information, route restrictions and fuel usage data are collected to detect operational anomalies and trends and then provide intelligent performance feedback on the findings.
Using Azure Stream Analytics and Microsoft Power BI dashboards, the goal is to uncover data insights that will enable airlines to improve their operational performance and increase fuel efficiency. And it’s on this last goal that Rolls Royce seems to be doing particularly well, with Microsoft’s head of Azure IoT, Sam George, detailing last week that the firm is saving $250,000 per plane by reducing fuel usage by just 1 percent.
Speaking here in London, Stein said that, in manufacturing, Industrie 4.0 is “a bigger deal than most people think”, with AI techniques enabling the firm to run a manufacturing factory which “starts to correct itself.”
The firm even has a case study of Industrie 4.0 – or the industrial internet – in action in the UK, and at its factory in Sunderland, which makes fan disks for jet engines.
The plant is using AI and machine-learning technologies to self-correct tool wear, as well as locate misplaced parts in the mill. “The results are quite remarkable – the machine can evolve and correct itself,” said Stein.
AI will see big benefits in manufacturing and automotive
Stein is excited by the future of AI in manufacturing, saying the “rich, digital connectivity between manufacturing and suppliers” will be an economic boon for Germany, China and “hopefully, the UK”.
“It’s going to be a huge productivity multiplier…it’s going to be a major international factor how quickly you can adopt [Industrie 4.0].”
For Medlock, whose SwiftKey firm was bought by Microsoft for $250 million in February, the future is about automotive and logistics. He says that the convergence of manufacturing vehicles and logistics through Big Data and AI could “reshape our physical world. “It would take out the need for parking spaces in cities, for example, and we could reclaim a third of the land we use to store cars.”
IoB has reached out to Stein and will update this story when we hear back.