What is the future of Industrial automation business (how will it grow) in this weak economic situation?

 

siva asked:

>What is the future of Industrial automation business (how will it grow) in this weak economic situation?<

Jim Pinto responds :

This topic is being discussed continually on my website:
"http://www.jimpinto.com":http://www.jimpinto.com

You might also look specifically at these articles:

New growth in old markets:
"http://www.jimpinto.com/writings/growth.html":http://www.jimpinto.com/writings/growth.html

2001: Industrial Automation Outlook (still applies):
"http://www.jimpinto.com/writings/outlook2001.html":http://www.jimpinto.com/writings/outlook2001.html

Further questions and/or comments welcome.

Cheers::

jim
----------/
Jim Pinto
email : [email protected]
web: www.JimPinto.com
San Diego, CA., USA
----------/

 

Frankly, I think Industrial Automation is one of those things that ought to be leading most economic indicators, not following them.

I hate to prognosticate, so instead, I'll tell you how I wish it would grow instead of how I think it will grow.

I wish it would take advantage of more open source software. Microsoft is asking for more license fees every time you turn around, and we're not getting as good a return on the investment as we did in years past.

I wish the industry would adopt some more useful language standards for control than Relay Ladder Logic. Using common languages such as C and C++ doesn't help much. What we really need is a scripting language oriented around time and event based controls and some control kernel utilities which clearly show latencies, pending tasks, memory usage, and other such things in a manner that doesn't require a computer science background to understand.

I also wish the industry would get serious with some pre-tested and packaged security configurations for SCADA and PLC networking. This would prevent the Operating System/Applications Software finger pointing hate that always seems to happen every time one tries to clamp down on security.

We need something better than the reporting packages and difficult to manage web browsing packages currently being sold today. Yes, part of the problem is that office applications are often ill suited for looking at real time databases. Meta-data solutions such as Verano's EnterpriseLink is a solution, but it's yet another layer to manage, and I'm uncertain that it would solve enough problems to make it worth while.

There! I've spewed some ideas. I'm sure others have theirs.

Hope you find your future in this industry...

 

The future is a very big place :^)

I think it will vary as a weak function of the economy and the cost of labor. As practiced today, it requires a lot of confidence that the expenditure today will really matter tomorrow. A small wavering of confidence tends to mean a big shutdown on spending. Watching the last
decade or so has shown the even a bear market didn't open the spigots wide. I think there needs to be lower cost/ greater value to produce much of an upturn. The rate of failed projects is discouraging as well.

In short, I don't see anything large and positive without a disruptive event or technology to put the conditions in place for a run up and/or generate some excitement.

Pretty much stagnant.

Regards
cww
--
Free Tools!
Machine Automation Tools (LinuxPLC) Free, Truly Open & Publicly Owned
Industrial Automation Software For Linux. mat.sourceforge.net.
Day Job: Heartland Engineering, Automation & ATE for Automotive
Rebuilders.
Consultancy: Wide Open Technologies: Moving Business & Automation to
Linux.

 

Regarding the future of Industrial Automation,
Curt Wuollet [[email protected]] wrote :

>I don't see anything large and positive without a disruptive event or technology to put the conditions in place for a run up and/or generate some excitement. Pretty much stagnant.<

Jim Pinto responds:
I AGREE (strong agreement)!

The industrial automation business is still selling PLCs (technology that is 30 years old, and a commodity) and DCS and HMI - all "old stuff".

The "inflection point" will come with new and significant technology. For my candidates & suggestions (too long to post here), please read:

What's HOT for industrial automation:
"http://www.jimpinto.com/writings/whatshot2001.html":http://www.jimpinto.com/writings/whatshot2001.html

Scarcity & Abundance - the Inflection Point:
"http://www.jimpinto.com/writings/inflection.html":http://www.jimpinto.com/writings/inflection.html

As an Automation engineer, don't wait around doing the same-old same-old stuff - PLC programming, SCADA and DCS systems integration. Use this stagnant period as an opportunity to move up to NEW business.

There is ALWAYS something NEW to get involved with!

Cheers::

jim
----------/
Jim Pinto
Tel : (858) 353-JIMP (5467)
email : [email protected]
web: www.JimPinto.com
San Diego, CA., USA
----------/

 

This is something where I tend to agree with Curt. Confidence is the big issue, and there is not a lot of it out there. Even the electrical
generation field, which was really hopping for sometime, now seems overdone, and there seems to be some scaling back, and postponements there.

I think we are still suffering from the Y2K thing as well. A LOT of resources were expended on dealing with this issue, and it may have come back to haunt us. A lot of people put in new systems, as they felt it was better than trying to patch up older stuff, but have yet to fully digest the new stuff.

The failed projects angle is also a real good point. I have yet to hear of a large automation project that actually managed to achieve its original goals. Many of them are just plain failures, at least financially, for various
reasons. The bean counters are scrutinizing things and thinking that nothing was really gained, so why spend more money? Smaller projects seem to be funded because its possible to show a return, but the larger, and more
ambitious ones seem to be slipping away.

Much of the failure seems to be related to the inability of people above the factory floor level to effectively use the information being generated by all the automation and SCADA systems now out there. The norm seems to be a lot of data, but it is so overwhelming that few are really using it effectively. I think this has led to managers no longer believing in automation as a way of increasing their bottom line.

Even when companies believe they gained something, they often are unable to quantify the gains. And promised savings rarely turn out to be anywhere near what was anticipated, so bean counters are discounting these numbers.

From what I can see, unless there is some other reason, most automation of existing equipment is not being upgraded unless there is an almost immediate boost to the bottom line. There are not many automation projects that can do this. The other side of this coin, is that a lot of existing equipment and processes are not well suited for automating. The time and money required to change the way things are done and procure new equipment far exceeds the automation costs.

One other issue is that the increases in productivity over the last few decades have been incredible. The amount of labor per unit of production has fallen to the point where additional improvement in that area is not all that easy. Waste has also been dramatically reduced so its difficult to gain a whole lot there either. Its hard to justify expenditures that only marginally impact labor or waste. Even if you reduce the labor content and waste to nothing, in a lot of products it would not matter a whole lot. I read somewhere that one of the car makers now takes only 24 hours of labor to
assemble a car. How much are you willing to spend to reduce it to 23 hours?

Bob Peterson

 

Unfortunately my view of Project Approvals or Expenditures, based on ROI (Return On Investment) or technical (it's the latest) merit, is somewhat
cynical. I worked for a number of medium, large, and extra-large companies. Authorization success was often achieved by:

a) Cost estimates covering major projects were intentionally kept low to make them more attractive.

b) ROI's were intentionally inflated.

c) Personal relationships between the person who wanted the project, and the approval board members.

d) Evaluation of a project's actual performance was seldom compared to the original "pie-in-the-sky" figures used to "prove" its effectiveness.

Well, I said it would be a cynical view!

Regards,
Phil Corso, PE
(Boca Raton, FL)

 

Hi Jim

That's a lot easier to think about when you're not battered and weary from trying to bring about even the most miniscule and rational changes. Or reviled for promoting disruptive technology.

Based on what I've experienced, folks are just gonna ride the spiral down until the innovation comes from outside and they're relegated to being mechanics and operators. Most people already just operate shrinkwrap software and buy ready made hardware. We're not very far away from being replaced with the man on the street. One more round of "ease of use" will do it. Dumbing down ends with extinction. It's the easy way out. Congratulations.

Regards

cww
--
Free Tools!
Machine Automation Tools (LinuxPLC) Free, Truly Open & Publicly Owned
Industrial Automation Software For Linux. mat.sourceforge.net.
Day Job: Heartland Engineering, Automation & ATE for Automotive
Rebuilders.
Consultancy: Wide Open Technologies: Moving Business & Automation to
Linux.

 

I think its unlikely that a specific "next big idea" (if it exists) would ever be posted here. If there is such a thing in the works I'm sure the creator of it is trying to develop it a bit more before tipping off the rest of us. Disruptive, radical, and revolutionary advances will likely take us all by surprise. That's why they are called disruptive.

For my money (as ignorant and meager as it is), it seems that this might come from areas involved in new computational theories involving complexity, neural networks, and cellular automata. Jim Pinto has been talking about this stuff in his newsletters. I haven't figured it out myself but reading about these things gives me the uneasy feeling that there is something going on here that I better learn about. I've read this book recently: "The Quantum Brain: The
Search for Freedom and the Next Generation of Man" by Jeffrey Satinover that, despite its title, is an excellent overview of these technologies and tries to tie them together with quantum theory and human free will. I'm also considering tackling "A New Kind of Science" by Stephen Wolfram. His web site, "http://wolframscience.com":http://wolframscience.com , has some intriguing excerpts. If you go to Jim Pinto's web site he has some links to reviews there that are interesting.

If half of what Wolfram claims is true turns out to be, this could turn most programmers into burger flippers (or some other heretofore
unknown profession) overnight. It could also increase wealth and productivity so much that burger flipping could be a real good job. Would you like some fries with that?

Regards,
Ralph Mackiewicz
SISCO, Inc.

 

> > I think its unlikely that a specific "next big idea" (if it exists) would ever be posted here. If there is such a thing in the works I'm sure the creator of it is trying to develop it a bit more before tipping off the rest of us.
>
> One possible exception would be Open Source; indeed, it's not likely to happen unless people are tipped off precisely on lists such as this.

An open source implementation of an almost 30-year old concept (PLC) is hardly disruptive. Useful? Yes. Disruptive? Not significantly.

> > For my money (as ignorant and meager as it is), it seems that this might come from areas involved in new computational theories involving complexity, neural networks, and cellular automata.

> Personally, I'd have more faith in sound engineering principles applied to programming than in neural networks; while NNs are useful for certain limited tasks, the current state of the art is sharply limited, for instance in the number of inputs and outputs they can handle. There does not appear to be a clear path toward large NNs, even in academia, which makes them pie-in-the-sky technology.

It's that pie-in-the-sky stuff that falls down out of the clear blue that ends up destroying entire industries. I'm not suggesting that everyone abandon sound engineering principles or that complexity based computations, neural networks, and cellular automata are ready to be used to replace current controls technology. Others may claim this to be true. But, I lack enough knowledge on these subjects to back up this claim. The question that was asked was about where the
future is, not where we are now. To assume that the only thing the future holds is more of the same is a dangerous assumption unless you are about to retire. Even then it may not be that safe if your retirement funds are tied up in the status quo.

Regards,
Ralph Mackiewicz
SISCO, Inc.
6605 19-1/2 Mile Road
Sterling Heights, MI 48314-1408 USA
T: +586-254-0020 F: +586-254-0053
mailto:[email protected] http://www.sisconet.com

 

Hi Ralph

Ralph Mackiewicz wrote:
> > > I think its unlikely that a specific "next big idea" (if it exists) would ever be posted here. If there is such a thing in the works I'm sure the creator of it is trying to develop it a bit more before tipping off the rest of us.
> >
> > One possible exception would be Open Source; indeed, it's not likely to happen unless people are tipped off precisely on lists such as this.
>
> An open source implementation of an almost 30-year old concept (PLC) is hardly disruptive. Useful? Yes. Disruptive? Not significantly.

The PLC is not the disruptive part. The vendor neutral platform is. Being able to work across and integrate product lines with OSS and Open Standards finally allowing best of breed selection, tearing down The Tower of Babel and ubiquitous Ethernet/TCP/IP networking would have far greater impact. And the greatest impact will be from intangibles. A community that shares knowledge and experiences, platforms that can be completely known and are fully and publicly documented and understood and most important, used and supported by more than one vendor or entity. And the dramatically reduced cost and flexibility will open up a whole new range of applications that aren't feasible now and are
becoming less so as costs skyrocket. What we have is a commodity market without the commodities. The potential for cost reduction if the industry were to converge on a single platform rather than
each and every vendor reinventing the wheel is staggering. The potential for progress if they weren't all working against each other is even greater. I would equate it to what happened after
IBM standardized the PC and before Microsoft wrecked the competitive mechanism. Having a platform designed for and controlled by the automation community would eliminate at least
half the problems and greatly improve productivity. If everyone didn't have to start at square one and repeat all the same mistakes and independently learn by trial and error, we'd build
a lot more solutions a lot faster and at a lot lower cost. The current model is the least efficient possible in many aspects.

> > > For my money (as ignorant and meager as it is), it seems that this might come from areas involved in new computational theories involving complexity, neural networks, and cellular automata.
> >
> > Personally, I'd have more faith in sound engineering principles applied to programming than in neural networks; while NNs are useful for certain limited tasks, the current state of the art is sharply limited, for instance in the number of inputs and outputs they can handle. There does not appear to be a clear path toward large NNs, even in academia, which makes them pie-in-the-sky technology.
>
> It's that pie-in-the-sky stuff that falls down out of the clear blue that ends up destroying entire industries. I'm not suggesting that everyone abandon sound engineering principles or that complexity based computations, neural networks, and cellular automata are ready to be used to replace current controls technology. Others may claim this to be true. But, I lack enough knowledge on these subjects to back up this claim. The question that was asked was about where the future is, not where we are now. To assume that the only thing the future holds is more of the same is a dangerous assumption unless you are about to retire. Even then it may not be that safe if your retirement funds are tied up in the status quo.

Similarly corporations that focus on maximizing profit and maintaining control to the point that they lose sight of market needs are vulnerable
to disruptive ideas that simply stress efficiency and cost effectiveness. Particularly if they are in the habit of shaking down their customers to
pay for their inefficiency. It's really simple, if someone can provide something that's better for your customers, lower in cost, and better
addresses what a total solution looks like today, you're eventually gonna have a problem. no matter how much you spend on marketing. Right now, people gladly pay 20X the intrinsic value if it's easy enough. When a commodity solution becomes easy anough, you're toast.

Regards

cww

--
Free Tools!
Machine Automation Tools (LinuxPLC) Free, Truly Open & Publicly Owned Industrial Automation Software For Linux. mat.sourceforge.net.
Day Job: Heartland Engineering, Automation & ATE for Automotive Rebuilders.
Consultancy: Wide Open Technologies: Moving Business & Automation to Linux.

 

Ralph Mackiewicz:
> An open source implementation of an almost 30-year old concept (PLC)
> is hardly disruptive. Useful? Yes. Disruptive? Not significantly.

Well, it might be disruptive to the folks that currently charge $$$$ a pop for SoftPLC packages

Another thing is that currently there is a great deal of repeated work in automation; there's only so many ways to wire a limit switch, yet each time the ladder is written anew. (No, cut 'n' pasting doesn't count.) Eliminating this repetition would free the programmer to do more useful things; and Open Source has as much chance as anything of doing that.

> > > For my money (as ignorant and meager as it is), it seems that this
> > > might come from areas involved in new computational theories
> > > involving complexity, neural networks, and cellular automata.

> > Personally, I'd have more faith in sound engineering principles
> > applied to programming than in neural networks;
...
> > There does not appear to be a clear path toward large NNs, even in
> > academia, which makes them pie-in-the-sky technology.

> It's that pie-in-the-sky stuff that falls down out of the clear blue
> that ends up destroying entire industries.

Normally, there's a several-year gap between academic and industrial. When the academic stuff starts looking hopeful, that's when it's time to
check it out. Anything earlier is premature.

> I'm not suggesting that everyone abandon sound engineering principles
> or that complexity based computations, neural networks, and cellular
> automata are ready to be used to replace current controls technology.

Sorry, that was a bit of a rant against an old Jim Pinto article, with NNs and similar technologies as the solution to the `too big' problem.

> To assume that the only thing the future holds is more of the same is
> a dangerous assumption unless you are about to retire.

Indeed; on the other hand, if it's about more distant future, indications are that within this century we'll get true nanotech and quite possibly human-level artificial intelligence.

Nanotech is intimately tied with automation - indeed, a lot of its (peaceful) uses will depend on pretty much classical automation, only done with large numbers of extremely small machines. From that point of view, it's more of the same - but it'll have such a huge effect on so many areas of human life that it's difficult to imagine coherently.

The effect of AI is much harder to predict. If Moore's law continues[1], the machine intelligence will outstrip us in pretty short order. The melodramatic name for this is `AI singularity', but the practical upshot is that it's basically impossible to say what'll happen after.

That's quite possibly within the lifetimes of many of us on this list.

Jiri

[1] If it doesn't, AI is much more doubtful.
--
Jiri Baum <[email protected]> http://www.csse.monash.edu.au/~jirib
MAT LinuxPLC project --- http://mat.sf.net --- Machine Automation Tools