VDC Research: Industrial Sensing, Networking & Control Market Blog

Predictions for Data Acquisition Solution Markets in 2012

With the last of eleven market studies covering the data acquisition (DAQ) solutions markets now being published I decided to take this time to reflect on what I learned over the last several months and to blog some informal predictions on what I may expect to see happen in the DAQ market in 2012 (admittedly I am not going out on many limbs with these).

1. 2012 shipments of external chassis & modules (i.e. DAQ systems), plug-in data acquisition analog I/O boards and standalone (non-bundled) data acquisition software will likely exceed $1.5 billion globally, with DAQ systems accounting for almost 76%.
2. Demand for modular data acquisition systems (one of eight platforms under study) will continue to grow at an above average market growth rate, comprising ~23% of DAQ systems shipments by 2012, as users desire the greater flexibility and ability to mix and match sensor inputs to meet their specific needs without spending on functionality which they do not require.
3. The number of data loggers designed with LCD displays will increase as other suppliers begin to realize that offering video graphic type functionality in a data logger is a sure way to gain market share in that category, just ask Graphtec, Keyence, Hioki E.E. and Dranetz-BMI.
4. Paperless chart recorder suppliers, for units with both >144 mm  and <144 mm panel cutouts, will continue to make inroads into non-process focused markets as an increasing number of these customers realize the benefits of greater real-time visibility into their operations, most of which are increasingly being automated through technology investments.
5. Demand for plug-in analog I/O boards will continue to lag that of most external chassis & module products as an increasing number of users seek out USB PC front ends as they are attractive as peripheral add-ons to computers that are portable and can easily be moved to another platform in a different location. Expectations in usage of USB 3.0 in all new PCs going forward is also expected to help fuel this trend.
6. There will be at least one merger and acquisition within the data acquisition solutions marketplace. The overall market is too large and too fragmented to not experience continued consolidation. Assuming that some predict the future by looking in the rear view mirror of history, some likely targets to consider watching may include Spectris (HBM and Bruel & Kjaer), National Instruments and DynamicSignals LLC. Who knows, maybe ABB or Honeywell will decide to get a little more active in the market and decide to expand their product portfolio beyond paperless chart recorders by acquiring a smaller player.

Okay, so maybe I decided to go out on a limb somewhat with the last prediction but that is what makes these things fun. I welcome to learn about any predictions you may have and/or what you think of mine.

A Few Thoughts on Solyndra

As the debate still rages on what happened with Solydra I thought I would weigh in with my opinion. I have never had the benefit of visiting that plant but have seen countless times the stock footage of President Obama's visit when things were going well. What you see in that footage was a lot of production cells with industrial robots moving solar panels through the various process stages. What you see in that picture correlates very well with what I saw when I did have the benefit of a detailed plant tour at Spire Solar in the fall of 2009.

Spire Solar is in the business of selling production equipment used in the manufacturing of solar cells/panels. Solar panels start with small cells that are tested and then progressively wired into larger and larger arrays in order to increase the voltage and current that is produced. Each circuit group is limited by the weakest cell in the group and so there are design trade offs to maximize efficiency and lower costs. The solar cell arrays are then connected together and encased in the panels that protect them from the environments they will be installed in. One last test involves exposing the newly manufactured panel with a calibrated amount of light that is evenly made across the entire panel. The panel output is carefully measured and the panel is graded. Some panels are much better than others.

So this leaves the question of what happened to Solyndra and why they could not compete with Chinese manufacturers and I think there are several factors that we should look at.

Industrial Automation: Solar Panel manufacturing almost demands automation. Manual processes risk contamination (think fingerprints) of the solar cells and weak cells mean less efficient and lower quality panels. Therefore, the labor head count and amount of automation should have been similar between Solyndra and its Chinese competitors. In an efficient process, the difference in labor costs should be a factor but not a show stopper.

Materials: This should be about equal between China and the US as the silicon materials are not exotic or in limited supply or any other global factor other than perhaps having a lot more semiconductor Fab facilities in the APAC region than the US.

Management: This I'm not sure about. How committed was the company to making solar panels as opposed to getting money and spending it under the guise of being a "green" business. I'll let the congressional investigation solve this one but it does seem clear that contributed capital was not efficiently applied to the facility and design/creation of the panels and processes to create them.

Power Costs: I am fairly certain that the cost of electricity to run a highly automated solar panel plant is significant and, costs are likely much higher in California than say China. This also by itself would not be a showstopper.

Machine Safety: Looking back at the stock video of the Obama visit to Solyndra you see that the automated robots have safety enclosures and features in order to bring them into compliance with OSHA and/or other standards such as EN ISO13849-1 and EN/IEC 62061. In our recent report on Machine Automatic Safeguarding we saw APAC as a smaller but growing market in that there is a lot of room for safety improvement - particularly in China. Machine Safeguarding adds expense and usually lowers efficiency to a sight degree. Let's consider this as a burdon for Solyndra that its Chinese competitors don't have and call it at that. A showstopper? I don't think so.

Economics: Two things are mainly at work here, a glut of solar panels and a decrease in demand as government subsidies dry out and the recent recession makes money tighter. Then, we have the fact that China is holding its currency down and that is a significant factor. If customers are buying on price alone and not taking panel quality and grade/efficiency into consideration this does put companies like Solyndra at a disadvantage.

In summary, I think it was not any one or two factors that caused Solyndra to fail. It was likely to be the combination of all of them happening at the same time for an extended period that did it. Certainly there are a number of valuable lessons to learn and business/economic factors that have to be corrected.

 

As Supporters of WirelessHart and ISA 100.11a battle it out, Are Users Sitting it out on the Sidelines?

There was a couple of different articles recently which discussed some of the issues inhibiting even greater wireless adoption (our networking research shows it already enjoys robust growth) in the industrial automation space. One article that really caught my eye discussed the impact that competition between Wireless HART and ISA 100.11a over which standard will be the dominant enabler of smart instrumentation is potentially having on user adoption rates.

Another reason the article was of interest was because it discussed many of the same hurdles which have previously been holding back greater wireless adoption and also the new technologies are also eliminating many of these same obstacles.

Although our recently published research only covered industrial wireless networking infrastructure products, which represents only a AMll share of products relying on Wireless HART and ISA 100.11a networks, our user survey respondents (comprised of end users, OEMs and channel) indicated a strong demand for networking products using both of these networks.

Expected Trend in Types of Wireless Infrastructure Networks Being Used in Industrial Facilities (Percent of Respondents)
Networks	Current                           Expected
	Share	in 2014
IEEE 802.11g	49%	29%
IEEE 802.11b	40%	31%
Proprietary in 900 MHz Band	37%	36%
IEEE 802.11a	33%	29%
Proprietary in 2.4 GHz Band	33%	35%
Bluetooth	32%	35%
IEEE 802.11n	31%	47%
Wireless HART	28%	40%
IEEE 802.15.4	17%	27%
ISA SP100.11a	15%	31%
ZigBee	6%	8%
Proprietary in 400 MHz Band	4%	6%
Proprietary in 800 MHz Band	4%	4%
Note: Percentages sum to over 100% due to multiple responses

The reason why the article really caught my eye was because it reminded me back in the day in 2001/2002 when I was covering the nascent markets for wearable technology and smart textiles. I remember speaking to some people over dinner, after a conference at which I presented on smart textiles, and I told them not to worry fighting over share position of a small, emerging market, and to focus on growing the pie (i.e. market) until it is big enough to be worth fighting over.

The arcane analogy is somewhat relevant since based on extensive discussions with 50 wireless networking suppliers the global share of wireless industrial networking infrastructure products for use in industrial facilities relying on Wireless HART and ISA 100.11a was 1.6% and 1.3%, respectively. In addition, this situation is frighteningly familiar to the wireline fieldbus network wars that I first covered back in 1997; some of those same wars are still being fought today at the Ethernet-based application layer network protocols in the wireline networking products market.

I am not going to debate the technical or commercial merits of either wireless network in this blog as the article in question does a decent job in covering the basics. I am also not going to publicly state which of the two networks I believe will eventually win out as we have clients in both camps.

Rather I am raising the question as to whether customers really are waiting on the sidelines for an eventual winner to emerge and what the potential impact will be on overall wireless adoption in industrial automation today and into the future. I welcome any thoughts or opinions you may have on the matter.

Storm Power Outages - Raise Tempers & Outcry

As the days drag on since last weekend's unusual October snowstorm these are still thousands of households throughout the US Northeast states without power. With each newspaper or TV newscast, you see stories of hardship and outrage over the delays in restoring power. Politicians make public proclamations about utility company incompetence and poor response and it's an easy target. While I'm pretty sure there are at least a few things that utility companies could do better, for the most part they are the doing the best that they can given the circumstances.

Pretend for minute that you are a utility company executive. You have stockholders to answer to and rates are often capped/regulated. Picture the manpower and equipment that you would need to provide the service that everyone is expecting in the wake of the storm. Everyone wants power back on within 24 hours if not sooner. The Northeast based utility companies have brought in crews and equipment from as far as Michigan and, as of today, they still have not finished. What does this mean? If you tried to have resources to deal with the situation in the manner expected, you would have to have at least 3 or 4 times the workforce and equipment. This means that for the other 95% of the year when there were no storms/crisis you would be paying 2/3 of that utility company workforce to do nothing. Electrical rates would have to go up greatly and then there would be a domino effect of consumer cutbacks and manufacturing jobs leaving the Northeast states for lower rate regions.

In closing, working with electrical grid high voltage lines and infrastructure has to be done in a controlled manner to avoid costly mistakes and delays. The crews are doing the best they can and pushing them harder or longer puts their lives at risk. Power companies do make plans and expand / share resources to deal with the seasons and anticipated weather events but this last one was so unprecedented, unexpected and widespread it would be impossible to ramp up for even if cost was not an issue.

I'm sure my view of this matter might be tested if in fact my power was out but give those utility companies and especially the crews you meet a break.

Data Acquisition Products Market is Consolidating

Based on soon to be published research VDC determined that the market for data acquisition products has not only grown faster than previous expectations, thanks in large part due to strong demand for modular data acquisition systems, but it continues to experience consolidation.

In 2008, back when the market was just over $920 million, VDC determined that the top 10 suppliers controlled just over 52% of the market for data acquisition solutions (external chassis and boards). New research indicates that the top 10 suppliers now control almost 55% of the $1.4 billion market in 2010.

The composition of the top 10 data accquisition solution (chassis, boards and software) suppliers themselves has also changed dramatically since 2008.

	2010 Shipments
1	National Instruments
2	HBM
3	Yokogawa
4	Agilent Technologies
5	Bruel & Kjaer
6	Campbell Scientific
7	Dewetron
8	Advantech
9	GE
10	VTI Instruments

In 2008 National Instruments was still #1 but most of the other suppliers have either changed position (only Agilent remained at #4), dropped out of the top 10 (Dranetz-BMI and Fluke) or completely exited the market as Sony did over a year ago. 

Companies such as VTI Instruments and Dewetron have made material gains in market share due to strong product offerings, strong technical support and savvier marketing. B & K was not covered in the 2008 research.

The size of the data acquisition market is materially larger than our 2008 research forecast due to expanded coverage which included modular data acquisition systems and standalone (non-bundled) data acquisition software as well as new coverage of VPX and XMC plug-in data acquisition analog I/O boards. Despite the expanded coverage the research indicates the core product markets have rebounded nicely since the global recession of 2009.

The modular systems category comprises the largest share among the external chassis & modules segment and is forecast to experience the strongest growth in demand through 2015. Clearly customers are demanding solutions which offer flexibility and agility in the ability to mix and match sensor inputs while also delivering required performance. Growth rates among VPX and XMC boards are also at well above average board market growth rates due to their enhanced performance, rugged design and growing popularity among military and aerospace applications.