Statistical Data Analysis

GE sees robotics, data shaping a new industrial Internet of Things

“We are on the verge of a new industrial revolution,” says Bill Ruh, GE’s vice president of software. The cliché is not lost on Ruh who has been ploughing this particular path since he first began working on sophisticated satellite and defence systems back in the 1980s. For so long robots and intelligent machine-to-machine communication (M2M) has been the stuff of fantasy or highly expensive R&D projects. But now this industrial Internet of Things could change all that.

Two years ago Ruh spoke to IDG Connect of a vision, a world where machines spoke to humans and helped them manage things more effectively. Today that vision is being realised with the emergence of what Ruh calls the “industrial internet” and GE, an international conglomerate that can trace its history back to Thomas Edison in 1876, is at the heart of it.

It’s a long way from the iconic 1968 Philip K. Dick sci-fi book Do Androids Dream of Electric Sheep? that spawned the classic movie Blade Runner, where robots and humans were almost indistinguishable. Yet Ruh and GE’s machine data analytics platform Predix are already laying foundations for a level of machine automation not previously seen - which is why, in Ruh’s words, so many businesses “are watching from the wings to see how it pans out”. The robots are definitely coming.

So what are these robots and what exactly are we watching here?

Not little metallic men; not yet anyway. ‘Robots’ here means automatons, self-operating machines that can learn from data analytics. It’s still early days of course…

“Robotics and drones, all that automatic tech will take everything up another level but you have to put the right infrastructure in place with the current machines in order to make that next-generational reality work,” says Ruh.


If you build it they will come

Ruh naturally thinks that Predix is part of that infrastructure puzzle. In his words, Predix is “an industrial grade data analytics platform, optimised to collect data in a dull, dirty and dangerous world of industrial machines.”

Surprisingly, GE has taken its lead from consumer internet services. Ruh speaks of being “jealous of the consumer internet giants” and what they’ve achieved in terms of collecting, analysing and managing huge amounts of user data. Predix, he says, is built in their image.

“We are envious of three things,” he says. “Firstly they [the consumer internet giants] deal with data and analytics at a scale - the industrial scales are exactly the same so we generate bigger mounds of data than most internet giants on a given day. We just haven't been able to collect and analyse it like they have. Secondly, they did it at a cost that was tremendously compelling over traditional IT spend. Thirdly, the ability to put out applications at speed and scale - they do releases in weeks and months not years. We said if we are going to be in this industrial internet world it's got to look like that. So we looked around and there isn't a platform that exists, so we built our own.”

Ruh cites some early wins. Adding sensors to wind farm turbines and analysing the data has already netted a five per cent energy generation increase. The system was able to identify how to improve turbine efficiency without adding extra costs. Attaching sensors to gas turbines and analysing that data resulted in a one per cent decrease in fuel used to generate the same amount of electricity which equates to billions of dollars of savings a year, says Ruh.

It’s difficult to ignore the potential here. As humans we are inefficient when it comes to crunching data. We need machines to work this stuff out for us although Ruh admits humans are still needed for reading data analytics results and reacting. But for how long?

“People can't absorb all of the information,” says Ruh, “so how do you collect data and find signatures? How do you find the people to do this that understand the machines, understand the potential issues and can analyse the data? That's a big issue.”

Again Ruh doffs his cap to the consumer giants.

“They’ve solved it,” he says. “They watch and collect everything and they paint a picture about you - they don’t do ‘alarms’, they do ‘understanding’.”


The rise of the robot

If you replace targeted advertising for targeted efficiency instruction you get the picture and he admits that ultimately this is going to be a machine-only world.

He says the big money is on two things; zero unscheduled downtime and resource efficiency.

“Neither sound sexy but they are in reality the two sexiest things because the amount of money to be made is incredibly high. If you can predict and manage to ensure you have no outages - no plane delay because of maintenance, no electricity outages, no disastrous oil pipe leaks, no products leaving factories with faults and therefore you’d have no product recalls... It’s transformative.”

‘Transformative’ is a word Ruh uses a lot but then he’s used to working on cutting-edge technologies that solve complex problems. Not that all industries are on board yet. There are he says a few early adopters but the nature of most industrial companies is conservative, shaped by years of heavy regulation and slow moving and expensive technological development. He’s convinced though that the efficiency results and costs savings shown so far plus the speed and low cost of implementation are going to be – yes you guessed it – “transformative.”

He has plenty of additional examples from connected lampposts in smart cities (GE is currently conducting a trial in San Diego, California) through to scheduling on railways but these are still very early days. Yes the revolution has started but it’s going to be a slow one. There are still so many regulatory as well as cultural issues to overcome including data sovereignty – a massive issue for multi-national companies – but he’s convinced the economics of it all will dictate in the end. And that means robots – lots of them.


Now read: GE’s Bill Ruh expects human-machine dialogue


« Predictive policing: Will the UK follow the US?


Zoltan Istvan: Science & tech could eliminate death by 2030 »
Marc Ambasna-Jones

Marc Ambasna-Jones is a UK-based freelance writer and media consultant and has been writing about business and technology since 1989.

  • Mail


Do you think your smartphone is making you a workaholic?