Big data means big advances for our quality of life, safety, and economy.
And it’s not a lofty goal. These ideas are real and happening now. Earlier this summer we talked about how we can now do real-time video analytics that make real-time facial recognition software possible and can also be used to orchestrate traffic lights to improve congestion. Last week, 3 major innovations were announced in Healthcare that could actually make the US Healthcare system work better than anything Obama and the government are doing in legislation.
And then there is General Electric.
Just four months ago, GE announced they were investing $105 million in Pivotal big data technologies to push their concept of the Industrial Internet. While market watchers paid attention, many may have thought that real results would be years off.
This week, at their second annual event Minds+Machines, GE sent a powerful message for big data to the world: its here.
Predix Reader via http://www.gesoftware.com/predix
To prove it, GE announced the availability of its new Predix big data software platform. They also talked about 14 new solutions based on Predix that are ready now, with paying customers no less. The solutions are set to save power, aviation, rail and healthcare companies $20 billion dollars a year, plus create new jobs and bring even more data science skills to the US job market.
These 14 new products are in addition to the existing 10 ‘Predictivity’ solutions they launched over the past year. For GE, this has meant an additional $290 million in revenues so far this year, with orders for $400 million. For the organizations using these solutions, the value is even more impressive. Two customers specifically were called out among the announcements this week:
- St. Luke’s Medical Center is using GE software to manage and analyze patient and equipment data, resulting in a 51 minute reduction in bed turnaround time and reduced patient wait times.
- Gol Airlines is using software to better track, analyze and adapt its flight routes and fuel consumption and predicts it will see $90 million in savings over the next five years.
To understand what Predix is, think of it as an platform for industries to centralize metrics through “The Internet of Things” in a data store that comes with a library of sophisticated analytic functions and is wrapped with a powerful visualization app. From a business perspective, it provides industries ways to get a massive headstart on harnessing the power of big data.
In a report published this week by GE, they predict its applications could boost GDP by $10-15 trillion in the next 20 years by saving labor costs, improving energy efficiency, not to mention the extra value derived from worker productivity gains.
Full description of these predictions are in the report. Two are obvious, reducing labor and increasing energy. Citing these 4 industries as only a portion of the market opportunity, GE submits that the maintenance and servicing of the machines that power these industries totals over 300 million man-hours a year at a cost of $20 billion a year. If GE is able to help these companies to be better connected to the data these machines can provide, they could help them to automate some routine maintenance and optimize service schedules to avoid downtime. The same big data that allows service schedules to be scientifically optimized also allows fuel efficiency to be addressed on a wide scale.
To appreciate what this means for the economy, think of GE’s gas turbine business. They service thousands of jet engines a year for airlines. If targeted maintenance could ensure these engines have optimal fuel consumption, they stand to save the airline industry billions a year. With the gas industry spending around $3 billion a year on gas, even a 1% savings is major for their profitability, never mind the environment. If you think about it, the same logic applies for any energy-dependent industry: automotive, oil & gas production, manufacturing, and transportation. GE stands to improve all of these businesses.
Now think about how we could even value accident avoidance. If jet engines are micro-maintained, the chances of an airplane crashes due to engine failure diminishes. How do you value the financial and societal impact of avoiding even one crash?
Productivity gains are also massive. Service technicians will not only better prioritize preventative maintenance, but with a blueprint of the most important service needs spelled out they have a chance at reducing the incidence of reactive service calls when machines break. Automation will also free the labor force to pay attention to higher order activities. Sure, some routine activities may be assumed by robots, but for managers and information workers, less time will be spent collecting and analyzing data, and more time will be used to address what insight that information gives you.
For instance, building service schedules will become a minor part of a service manager’s day and instead they can spend time on hiring, training, and quality control. Similarly, instead of spending weeks collecting data and setting it up to analyze it, quality assurance engineers can assess real-time data and determine if a dangerous product defect exists in hours or days, gaining time to address the issue and correct it before it costs them or their customers money.
With that in mind, I think we should look at GE’s predictions of $10-15 trillion in savings as conservative, and start looking far and wide on what other opportunities there are to realize their vision of the Industrial Internet.
And while you’re at it, check out Pivotal HD and HAWQ which are Pivotal’s big data products based on Apache Hadoop that make the technology advances talked about here possible.
- Read the full press release which itemizes the following 14 new products, among other news of expanded partnerships with AT&T, Cisco and Intel.
- Read VentureBeat’s coverage on how this won’t affect employment rates
- Read more about Pivotal’s partnership with GE
- Read more about Pivotal HD and HAWQ, Pivotal’s distribution of Apache Hadoop that improve performance of Big Data queries by 318x for one retail customer
About the Author