The basic components of the Industrial Internet encompass intelligent sensors and instrumented industrial machines, accessed via networks that provide high-level data visualization and use sophisticated software applications to provide advanced analytics. (Source: General Electric Co.)
GE's report defines the potential of Ethernet connectivity, but leaves out one painful reality:
In a perfect world, all of that data would be easily accessed and that Trillion Dollar expansion of worldwide GDP could be realized. In reality, hackers penetrate systems just for the fun of it. Stuxnet leaves one wondering about government level penetrations as well.
Al, I want to start out to say that I am impressed by what GE is doing. I reported on some aspects of it when I wrote an article for this site on the Cloud Computing conference a few months back.
One of the things about Big Data that is seldom mentioned is the reason for the emphasis on it now. The reason is simple. Storage is cheap and networking is ubiquitous. Big Data is really just a new buzzword for data mining. It is in the mining of the data that the benefits come. Very large data sets and data mining have been around for a while. I worked for both Oracle and IBM in the database technology field. Over ten years ago we were already talking about, and helping customers with, petabyte databases. The large databases spanned the application field from retail (e.g., Wal-Mart) to nuclear weapons. What has changed, as I said, is that storage has become cheap, so now a lot of data that was not worth storing at previous prices of storage is being saved. It is basic feeds and speeds.
The real question that one has to ask about any investment is what is the Cost vs the Benefit. Frankly, in one area of GEs expertise, electricity generation, there is a real question of the benefit of all this. Without Big Data, or any of the other industrial internet stuff, electricity use growth in the US is down to almost zero (last number I saw was 0.8%). At the same time we are using more and more electrical devices. What is going on here. The plain fact of the matter is that the devices are getting more efficient. Computers and computer control of devices is a big part of it. Electric motors have been getting more efficient for decades (I used to follow that in the in-house research journals when I worked at GE, a couple of decades ago). Besides these efficiency increases, there was the increasing cost of the raw materials of energy. None of this required Big Data. The fact is that electricity is, for most of us, not the most expensive thing in out lives. Food and transportation take a lot more of our income. So, we could halve our use of electricity while not making much of a dent in a family's expenses.
I could go on and on about ideas that seem good, but that are not economical. Sufice it to say that, for this to be wotth while, it needs to be a good investment. If it is, it will be done. It already has in areas where the equation holds. As the components of this type of infrastructure get cheaper, the use will spread. To make this into a grand scheme, though, I think will be a hard sell.
TJ, The security problem isn't going away but all manufacturers will need to raise their game in that area if they want to reap the benefits of advanced networking. I read recently that studies have now shown that most attacks occur from within the intranet and not, as one might assume, from external internet connections. That creates a whole different bag of snakes to contend with in terms of disciplined internal connection to the plant network. Good point.
Is complexity coming to a point where it's too difficult to handle anything? Or is this just another set of infrastructure that will be handled by automation? For example, hand cranking cars, manual elevators, etc. Their replacements are very complex, but we take them all for granted now. I will assume GE's endeavor will succeed. They obviously see where automation could help weary companies.
The Utility/Power companies are too diverse in legal requirements and ownership to see much investment in the "internet of things" .... yet.
This type of technology will be used by corporations involved in real competition, not power / utility companies with profits set by local / state governments. To be really useful all the players have to "sign on", not just some.
As to security.. it isn't really a tech problem ... it a people (operators) problem. We have been involved for some time in this industry.
1) Getting SCADA units involved in control of power sub-stations back for repair... with original passwords still being used! Even simple levels of protection are ignored.
2) A friend was demonstrating remote monitoring and control (over internet) of a large gas turbine generator using our controls.. was supposed to be a simulation to a customer. For 6 months he was powering up/down a REAL generator during his "demonstrations" (yep ... generator was on-line during this period)
Actually, I believe we are quite close to a tipping point in the Internet of Things. The "app" universe has given us a great model of how Internet-connected devices can be deployed ad-hoc and form all kinds of mesh networks in the social realm, like Foursquare, Waze, and Twitter, etc.
This is not a product push, but I've been teaching and writing about Jeff Hawkin's thoughts "On Intelligence" since I reviewed a pre-release copy of his same-titled book in 2004. Working through his framework of human intelligence has resulted in a fresh approach to the analysis of "Big Data"... namely, "Big Data" is great for producing a documentary, but he suggests that real-time, in-line, on-line, just-in-time (pick a phrase) analysis of multiple data streams it the real linchpin to producing automated intelligent networks. No need to mine, just pan out the flowing stream (to continue the metaphor).
With the confluence of fast data networks, multicore processors, parallel GPU processing, and low-power networks such as Zigbee, 2013 just may see some smart networks arrive that can start out small and grow, rather than requiring the buildout of a grand plan...
DN has pointed this out before...educated engineers! As soon as you say the word 'internet' most college students dream of making it big via a 'Zurckerberg" breakthrough. Not exactly data mining but developing the next social fad!
Cabe, I do think the problem will be solved, or advanced depending on how you look at it, through software that deals with all of the complexities that the Industrial Internet would present if applied on a large scale. Smart networks could definitely provide a technological push but there is also a need for application software that translates the technology possibilities into a business case. I would expect that is GE's view -- their ability to provide enterprise software that provides management of key resources that rely on smart sensors, for example, to add field intelligence to these systems. It's going to happen, just a matter of when.
You got it, GTOlover. What amazes me are the resumes that come in with skills such as "social media". Now if these were marketing people who interned in the a social media space, that would be one thing. Unfortunately, I don't have much choice than to interpret that as "Facebook and Twitter Expert".
Truchard will be presented the award at the 2014 Golden Mousetrap Awards ceremony during the co-located events Pacific Design & Manufacturing, MD&M West, WestPack, PLASTEC West, Electronics West, ATX West, and AeroCon.
Robots that walk have come a long way from simple barebones walking machines or pairs of legs without an upper body and head. Much of the research these days focuses on making more humanoid robots. But they are not all created equal.
The IEEE Computer Society has named the top 10 trends for 2014. You can expect the convergence of cloud computing and mobile devices, advances in health care data and devices, as well as privacy issues in social media to make the headlines. And 3D printing came out of nowhere to make a big splash.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.