Automation Market - Real Numbers

L

Thread Starter

Luis Andes Navia

I just started at the beginning of this month, and is really impossible for me to find reports about the automation market. Reports that give me an idea about: -With each the most important brand in ......(PLC, Safety Relay, Robots ..). -Which is the situation in diferents countries (USA, Germany, Japan, ..). -Which is the opinion of the final and medium customer about the new software for this industry. -Etc.... In comparison with other this it is a market very hidden (in terms of information, reports,etc). Thank you very much for your opinions and for your help?
 
W
There are several companies who publish for sale reports on the automation market. ARC, ARM, Gartner Group, Frost & Sullivan, Venture Development Corp. and others. Copperhill & Pointer is publishing a series of reports on flow sensors. They all have websites. Also you can purchase individual pages from these reports at Northern Light. The automation market is not hidden at all. You just have to pay for the information. Walt Boyes --------------------------------------------- Walt Boyes -- Telesian Technology Inc. [email protected] 21118 SE 278th Place - Maple Valley, WA 98038 253-709-5046 cell 425-432-8262 home office fax:801-749-7142 ICQ: 59435534 Telesian Technology, an e-Business services firm specializing in industrial enterprises. http://www.telesian.com ------------------------------------------
 
C

Curt Wuollet

Hi Luis I have attempted to do market research on the industry too. Not only is it the most closed industry I have ever worked in, the information I do get is often obviously conflicting or impossible. It is even quite difficult to simply ascertain who is leading in a sector. There is considerable information in the archives here and on Jim Pinto's site, I believe. Good Luck! regards cww
 
C

Curt Wuollet

Hi Walt I guess nothing is closed if you have enough money. You could even buy the formula for Coca Cola. I guess that means it's not secret. Regards cww
 
J
In all likelihood, ARC Advisory Group (www.arcweb.com) has the numbers on the automation industry that you are looking for. If you want them, it will cost you as that is how we make a living, but they are good. Virtually all automation suppliers use our numbers for their planning purposes. John Moore Vice President ARC Advisory Group phone: 781.471.1151 http://www.arcweb.com/ Collaborative E-Manufacturing Forum 2001 June 25 & 26, Boston MA http://www.arcweb.com/arcweb/Events/Forum0601/forum0601.htm
 
K

Kerry L. Schrank

You don't need to do market research for the prevalence of PLCs in the industry. PC based control has taken over... :) Regards, Kerry Kerry L. Schrank Entivity 8631 Sea Ash Circle Round Rock, TX 78681 512-733-5811 tel 512-733-5806 fax 512-750-7662 cell www.entivity.com www.thinkndo.com
 
W
No, you can't buy the formula for Coca-Cola. The six people who know it are not even permitted to be on the same airplane. But you can buy reports on the state of the automation market. This is a business, just like any other. David Spitzer and I are producing a report on the magnetic flowmeter market...an innovative one because we've also produced a technical comparison between every model of meter made. It will take several hundred hours of our time to produce this report. It costs what it costs because we need to be compensated for our time. You are an engineer. What would you charge for a project that took, oh, say 300 hours to do? Dave and I, Jesse Yoder, Steve Walton, ARC, ARM, Gartner, Frost & Sullivan, Venture Development and quite a few others all produce reports for this industry. Don't say the industry is closed and no information is available, because it is not true. You can even buy specific pages from some of these reports from Northern Light. What you are complaining about is not that the information isn't available, but that it isn't free. If the entire world will pay me $250K USD per year, I will make all of the automation information you could ever want free and in public domain. Otherwise, you will allow me (and the other people who act as industry analysts) the right to earn our livings producing this information. "Information just wanna bee free." Batpuckey. As a net producer of information, I am perfectly willing to not give it to you unless you pay me. Am I holding information hostage? Heck no. I am charging for what I know how to do, just like you do. Should engineering be free? Part of what I do is to organize the data. Data is not information. Data is for whoever wants to collect it. Collecting it takes time and costs money. "Data just wanna bee free!" is batpuckey too. It is like the old story of the boiler engineer who comes to the building, looks at the instruments, gets out a hammer and bangs three times on a pipe...whereupon everything works perfectly. He sends a bill for $500. The facilities engineer protests, "all you did was tap on a pipe!" The boiler engineer says, "Okay, here's a revised bill. $3 for the pipe-banging. $497 for knowing where to bang, and why." Walt Boyes
 
C

Curt Wuollet

Hi Walt And those are the conflicting, and in some cases obviously impossible ones I referred to. They should at least give a preview so I could buy the figures I like :^). Did you know there are three leaders in the same area? That is Bullpuckey. I have settled for a composite of the views of folks on this list who, in my opinion have the least vested interest. Don't feel bad, analysts in the computer industry put out very predictable and conflicting reports based on who commisioned the study. That's where I learned the value of analysis done to order. One in particular you mention Gartner, produces the most obvious propaganda in the computer industry as well. I will revise my statement. Information is easy to find. Credible, objective, information that I dare to steer by has eluded me and paying for it didn't remedy the situation, it merely increased the variance. That still passes for closed in my book as it is impossible to verify. And I don't begrudge you whatever you can get people to pay for that information. I make a living knowing things. Regards cww
 
L

Luis Andes Navia

Sorry Kerry, but you know that is not true ........ yet. Go with your Ideas for a real time control automation system and will see what the client said to you.
 
L

Luis Andes Navia

What is the meaning of "Virtually all automation suppliers use our numbers for their planning purposes".... The market is hidden. You can make an analysis since the point of view of customers, technologies, marketplaces, competitors analysis,... but what about real numbers???. The automation companies (like in some others markets) never show real numbers about products. In fact it is very difficult to analyse the market becouse like you said are "your numbers" not the companies (ABB, AL, Simens, Schneider, Omron, Br, Lg, etc..) numbers. Best regards, Luis Andes > In all likelihood, ARC Advisory Group (www.arcweb.com) has the numbers on the automation industry that you are looking for. If you want them, it will cost you as that is how we make a living, but they are good. Virtually all automation suppliers use our numbers for their planning purposes. <
 
M

Michael Griffin

Would you care to show us some market research figures which support this statement? And they better be nice open figures too - not those nasty conflicting or impossible ones! ********************** Michael Griffin London, Ont. Canada [email protected] **********************
 
H

Heavner, Lou [FRS/AUS]

Luis, I'm not in a position to debate the accuracy of ARC's or anybody else's numbers. But I believe they are able to get data from automation companies that would never be shared directly with competitors or the public in general. The reason is that each company's specific numbers are held confidential and only aggragate market or market segment numbers are published in their reports. The more trust that automation companies have in a market analyst to protect their proprietary business information, the more they will be willing to share. Can an automation company lie about their numbers? Sure, but I suspect the analyst companies use numerous methods to validate their data. They probably do things like statistically sample automation buyers and check public information like automation company annual reports to at least ensure consistency. And they do this year after year. Most automation companies and potential buyers of automation market reports are probably more interested in trends than in the absolute accuracy of the data, anyway. If the information and analysis provided by the analysts was not of reasonably good quality, I don't imagine it would be as widely sold for as high a price as it is. And remember, just as automation companies compete, so do market analysts. So the prudent customer doesn't get "locked in" to one supplier unless he is confident that the supplier will fairly and adequately meet his needs making additional reports merely redundant. The various reports and surveys act to validate one another. The relative price of each report, all other things equal, is a proxy for the quality of the data. Regards, Lou Heavner Emerson Performance Solutions
 
J

Joe Jansen/ENGR/HQ/KEMET/US

Actually, I had assumed he was joking! ;^) I don't know about anyone else, but I have been hearing of the demise of PLC's for over 10 years, and yet *somebody* is still buying them; and paying me to write programs for them! --Joe Jansen
 
M

Michael Griffin

If you listen to computer people, they will tell you that the PC is going to disappear. It will be replaced in most applications by thin clients and a hoard of specialised gadgets connected to the internet. Certainly all the major office software companies are planning their future products on this basis. The future is supposed to be that you plug a thin client (you can think of this as a dumb terminal with a web browser built in) into the internet, and access all your applications from an "applications service provider" (ASP). You won't buy most software anymore, just rent it from the ASP. Customer companies will outsource all their current office computer operations to ASP companies. This is just like the old time sharing systems that used to run on mainframes. A similar market will serve the typical home user (most of whom only want web access and e-mail anyway). If this ever really happens, the mass market for PCs will disappear. There will be just thin clients, big servers, and "engineering workstations" for those oddball applications (CAD, software development, etc.) which don't fit the ASP model. If there is no mass market for PC hardware though, where does this leave PC based control? Who's going to build the commodity hardware? The industrial PC manufacturers depend upon being able to operate in parallel with the office computer market to give them a wide component supply base. Thin clients and large servers may not be suitable hardware models to base industrial PCs on. The only thing I am willing to predict is that most people who make forecasts about technology are going to be wrong... ********************** Michael Griffin London, Ont. Canada [email protected] **********************
 
B
My personal opinion is that this whole idea of the "thin client" thing is just hype from the software people who want a more assured revenue stream and the D/P departments who see it as a way to reassert control over computing resources. My guess is it not going to happen. The cost for a thin clinet terminal is not much different then a cheap PC clone with its own hardware. The software cost is rarely that much different either. The only place where they can claim any cost savings is in the "total cost of ownership", which is highly inflated to (IMHO) justify the D/P department's attempt to reassert its control over computing resources. This may, or may not, fly. My guess is that in highly centralized, and inefficient companys, and in a few applications, it may well become common. For most of us, its a bad idea that needs to be killed off.
 
J

Joe Jansen/ENGR/HQ/KEMET/US

Michael Griffin <[email protected]> wrote: >If you listen to computer people, they will tell you that the PC is >going to disappear. It will be replaced in most applications by thin >clients and a hoard of specialised gadgets connected to the internet. &lt;sarcasm> Does that mean that in a few years, after the PLC has gone away, of course, that my industrial controls will actually be running on a remote host and sending I/O data in XML via JavaScript? This will, of course, all run within a browser, correct? <&lt;sarcasm> Note the XML-Style tags!! I'm Gettin' Ready!!!! --Joe Jansen Future maintainer of the "industrial-ASP-Javascript-Remote-I/O archive"! dot-com
 
M

Michel A. Levesque, eng.

> If you listen to computer people, they will tell you > that the PC is > going to disappear. It will be replaced in most applications > by thin clients > and a hoard of specialised gadgets connected to the internet. > ....... It seems to me that we are closing the circle here. Let's think back to yesteryear when all computing was done on the "big iron" mainframes and dumb terminals. Then the PC revolution came about and gave everybody the ability to do their own thing (empowerment). Now we have the IT people trying to bring back the centralized "big server" mentality with limited power "thin clients". Seems like a power-grab to me. In automation, we saw the same thing..."big iron" DCS versus empowered PLC's. Now that PLC's are almost up to the DCS in terms of what can be controlled, will we see a power-grab from DCS manufacturers??? Let's face it, having "big iron" is an advantage for MMI, data acquisition, advanced controls etc. But I would stay with the distributed model for control (many PLC's interlinked). The centralized model has benefits for ease of maintenance and troubleshooting. But, it also has a common point of failure. Using "big iron" servers with thin clients can be done now (RSView is TSE compliant isn't it?). So why are we not jumping whole hog into this???... Because, if that one server goes down, you are up the creek without even a boat, let alone a paddle. Sure, we can set up redundant servers redundant networking etc., But wouldn't we be right back to distributed computing? I've been going around and around on purpose. I just wanted to make my point that fundamental changes like distributed computing and centralized computing have both been tried. We need another alternative (other than the traditional distributed or centralized) computing model in industry. Now, THAT would make money (manufacturers...are you listening?). Michel A. Levesque eng., mcp Directeur Bureau Montreal AIA Automation Inc. [email protected]
 
M

Michael Griffin

At 11:53 20/04/01 EDT, [email protected] wrote: >My personal opinion is that this whole idea of the "thin client" thing is just hype from the software people who want a more assured revenue stream and the D/P departments who see it as a way to reassert control over computing resources. > Yes, the software companies want it because they want to charge you a regular "rental", instead of just selling you the software. Customers are starting to resist buying upgrades they don't need, and the software companies are worried about that. Also, the current fad among stock analysts is that there is more profit in services than in products. Companies are trying to promote themselves as "services" firms so they can curry favour among these analysts. I think you are wrong about the D/P departments though. Under the original "thin client" model, they were going to provide and control the applications. The current business model though eliminates them completely, with this function being out sourced to ASPs (Application Service Providers). Many of the current software publishers (including the very biggest ones) want to reposition themselves as ASPs, or as service providers one tier above the ASPs. It is the possibility of outsourcing all the DP or IT functions, rather than any technical merit or actual cost savings which might drive this process. >My guess is it not going to happen. The cost for a thin clinet terminal is not much different then a cheap PC clone with its own hardware. The software cost is rarely that much different either. > >The only place where they can claim any cost savings is in the "total cost of ownership", which is highly inflated to (IMHO) justify the D/P department's attempt to reassert its control over computing resources. > I would expect that in real volume manufacture, the hardware cost would be about half that of a PC. There won't be any software capital cost, because the software will be rented - you will pay for usage. The big savings is supposed to come from eliminating your DP or IT staff. Arguing whether this will really save any money or not is moot. If the bean counters like the concept, they'll cook up some numbers to prove that it saves money. Large businesses today are run according to fads, not facts. > >This may, or may not, fly. My guess is that in highly centralized, and inefficient companys, and in a few applications, it may well become common. > >For most of us, its a bad idea that needs to be killed off. I agree with your last sentence - for most of *us*, its a bad idea. Most of the people on this list do the sorts of things which don't fit the ASP model. They will still need something which is recognisable as a PC. The limited market for these PCs (or workstations) means that they may be a lot more expensive than they otherwise would be. For people who only want to write memos and send e-mail though, this is less of a problem. Most of them probably wouldn't be able to tell the difference between a thin client and the PC they used to have. Joe Jansen wrote: >Does that mean that in a few years, after the PLC has gone away, of course, >that my industrial controls will actually be running on a remote host and >sending I/O data in XML via JavaScript? This will, of course, all run >within a browser, correct? Now we get to the meat of the issue. No, what it means is that if the above scenario comes true (and I am not sure that it will), then PCs won't be as common, or as cheap as they are today. Thin clients and servers probably won't be a good hardware model for small industrial controllers. Where does this leave us then? Well I bet that someone would come up with the idea of mass producing inexpensive proprietary hardware intended specifically for controlling industrial machines. It would come with the controlling firmware already installed. You would just need to write your application off line, and load it into this "programmable controller" via a serial cable. You might even be able to buy a controller with your I/O already built in. What a novel idea! Given some of the things we have heard lately, I bet that someone would even be able to patent the idea. The real point is not about the details of how well the ASP concept will work. The point is that the PC market will follow its own logic, without reference to what our needs in industry are. Mainframes went obsolete. Minicomputers went obsolete. There is absolutely no reason why PCs can't go obsolete in their turn. All that is needed is something to replace it. This "something" need not be evident to us at this time. Need I point out that in the mid to late 1970's it was not obvious that personal computers would be anything other than toys for hobbyists? If this all comes to pass, then perhaps PC based control using common office computer hardware and software (e.g. Windows) components will turn out to have been nothing other than a brief passing fad that never amounted to much. Industrial computers would still be around, but they would likely derive their hardware and software foundations from sources different from what is used in the office. I am not suggesting that any of the above will happen. I am just saying that it might happen. When we start forcasting the future of the automation industry, we need to keep this in mind. ********************** Michael Griffin London, Ont. Canada [email protected] **********************
 
C
Actually it's quite interesting that unless you were chained to Microsoft we have always had thin clients. It's just that in the UNIX world they are called X Terminals and the capability comes with any graphical UNIX system. Linux on a minimal PC makes a great X Terminal for the cost of the PC.. The version of Cimplicity IU that we have supports them. It's MS replacement does not. Now, I imagine the latest and greatest thing will be expensive thin clients that need both a Windows license and a seat license. Maybe a Citrix(or somebody) license too. Now if GE would just dump their crappy Windows product and support what was Cimplicity IU on Linux they would be right up with times and have a superior product with inexpensive thin clients. While they abhor cheap anything, I'll put it this way. They could charge what they get now and the customer would pay less. But one wonders why they were not a hot idea, apparently, then. What has changed to make them a hot item now? Regards cww
 
D

David McGilvray

Hi Michael, A couple of points you may wish to consider: 1. Modern server technology spans quite a range, from dirt cheap PCs, to fully redundant machines that MAJOR corporations are running mission critical apps 7/24 with the same financial downtime consequences as industrial applications. The advantages offered by the server option, compared to PLC/MMI or DCS alternative, are many but I'll mention a few: i) much greater flexibility (solution to fit application), ii) much richer development environment without the endless bottlenecks created by PLC and DCS hardware interfaces, contributing to superior systems and performance, iii) economies of scale, iv) greater operational efficiency when, like the internet, a client (thin or otherwise) can access the server, without concern for location or machine (reduce sneakerware). 2. Many server based systems now offer embedded system (with a number of operating system options including embedded linux, RT Linux, CE, Embedded Java, RTOS, etc) capability so that actual control code may physically operate in field distributed devices connected to the server transparently via a variety of fieldbus options. This gives a new meaning to the word distributed. dwm
 
Top