• Font Size:
  • A
  • A
  • A

Feature Articles

Machine Vision Market Outlook for 2005

by Paul Kellett, AIA Director of Market Analysis - AIA

What will 2005 bring for machine vision companies? Will it be a good year?  Or will sales flatten and even decline from 2004 levels? 

To answer these questions, we examined macro-economic and sector forecasts and queried leading machine vision industry experts for a more specific perspective. 

We asked these experts about their expectations for the machine vision market as a whole and where the largest opportunities would lie, zeroing in particularly on China.  We also asked them about what major market trends would emerge.

What did we learn? The upshot is that 2005 looks like a solid year for machine vision sales in North America, although at a level somewhat below that of 2004 in terms of percent growth.  The picture for worldwide machine vision sales, however, is less clear.

To arrive at this finding, we began with macro-economic and sector forecasts.

Macroeconomic and Sector Outlook
The macroeconomic outlook for 2005 is mixed, depending largely on region.  For the world as a whole, real GDP growth is expected to slow from 4% in 2004 to 3% in 2005, according to the Economist Intelligence Unit. However, actual performance in 2005 could drop below the anticipated level due to a number of disruptive factors including another sharp spike in the price of oil, a large increase in interest rates internationally, destabilizing trade imbalances and their effects on currency exchange, as well as the risk of a sharp downturn in the Chinese economy.

Annual OECDAccording to the OECD (Organization for Economic Cooperation and Development), the recovery of business investment should continue in North America and start in earnest in Europe in 2005, supported by strong balance sheets and high profits. While the “momentum of the recovery will benefit from continued Asian dynamism, …it remains to be seen whether Continental Europe will play a strong supportive role through a marked upswing of final domestic demand.”  As shown by Exhibit 1, the OECD’s forecasts of annual real GDP growth for 2004 – 2005 reveal declines for the USA, Japan and total OECD (Europe, Canada, USA and Japan) but a very slight increase for Europe.

OECD Forecasts of Real GDP GrowthExhibit 2 displays the OECD’s quarterly forecasts of real GDP growth. These forecasts show a dip in real GDP growth in 1st Quarter 2005 from last Quarter 2004 for the U.S. and Japan, but while the U.S. steadily improves thereafter, Japan improves in the 2nd Quarter and then levels off.  By contrast, total OECD countries are projected to achieve a steady improvement in the rate of GDP growth through 3rd Quarter 2005 and then level off.  Finally, the Euro Area is expected to display a growth pattern similar to that of the total OECD, albeit it at a lower level and with improvement coming in the 1st Quarter.

The outlook for GDP growth in the U.S. is favorable according to all of the forecasts we examined. According to the December 2004 CEO Economic Outlook Survey, “America’s leading CEOs expect the U.S. economy to continue to grow at a healthy pace in the first half of 2005, with a slight easing from strong growth in 2004.”  The CEOs are assuming 3.5% real growth in GDP, down from 4% in 2004. 

Wharton economists see a similar pattern for the U.S. economy.  According to Wharton finance professor Jeremy Siegel, “I think we can have 3.5% to 4% real growth (in GDP) with very moderate inflation, 2.5%.”  Wharton finance and economics professor Richard Marston says he too expects GDP to grow 3.5% to 4% in 2005.  This growth, however, could fall victim to a number of threats. A weakening dollar, fueled by the federal budget deficit and current account deficit (balance of payments), could fall even more if foreign investors start selling US treasuries. Per professor Marston, “the current level of foreign financing of U.S. debt is not sustainable.” To attract investors, treasury bond yields would have to rise, causing interest rates for all types of loans to increase, dampening U.S. economic growth. Also another surge in oil prices would increase business expenses, undermining profits and thus depressing GDP growth below anticipated levels.

Real GDP Annual Growth in ChinaSince China represents the second largest economy in the world in terms of Purchasing Power Parity (although seventh in terms of GDP), and because many North American and European manufacturers view it as a growing market opportunity, we also examined GDP forecasts for China. According to the Economist Intelligence Unit, China will once again be the fastest growing country in Asia in 2005 with real GDP growth over 8 %. The OECD also projects real GDP growth in China at 8% in 2005, a decrease from 9.2% in 2004.  According to Bloomberg’s New Service, China’s economy will probably expand in 2005 at its slowest pace since 2001 but still much faster than elsewhere. Based on a median forecast of 12 economists, Bloomberg believes growth will probably slow to 8 % from an estimated 9.3 % in 2004.  This growth pattern is depicted by Exhibit 3.

Representing a less sanguine view, the former head of the World Bank office in China, Yukon Huang, said the country faces severe sustainability challenges, especially regarding oil and water.  He predicts that China’s economic growth, which was 9 % in 2004, is likely to fall to a more moderate 6% to 7% in coming years, still robust growth compared to other countries. 

Based on data availability and importance to the machine vision market, we next examined forecasts for the semiconductor sector. What we found is a mixed picture. According to the SIA (Semiconductor Industry Association), worldwide sales of semiconductors will be flat in 2005. By region, sales performance will vary.   In the “Americas” market, a 4.5% decline in 2005 is expected.  In the European market, no growth is expected in 2005 over 2004.  In the Japanese market, minimal growth of 0.2% is anticipated, and in the Asia-Pacific market, modest growth of 1.9% growth is predicted. In-Stat, a market research firm, takes a more dire view; according to this research firm, 2005 semiconductor sales are expected to decline by 5.7%.  This is very close to SEMICO’s forecast of –4.7%, which was presented at the SEMI Industry Strategy Symposium in January, 2005.

A still more divergent view is offered by VLSI Research, which expects 3.8% growth, albeit with some deterioration towards the end of the year.  Their reasoning:  “For a downturn to occur, normally there would need to be excesses such as higher inventories, declining IC (integrated circuit) prices, and declining RPSI (revenue per square inch produced). These conditions are currently either minimal or nonexistent. Inventories have declined since the summer, and IC prices have remained strong as 2005 begins. Moreover, the RPSI has steadily increased throughout 2004.”

What about the semiconductor market in China? China’s integrated circuit market is forecast to grow 11% in 2005 according to IC Insights. In 2005, China is forecast to represent 20% of the world’s $175.4 billion in IC consumption. This, however, does not mean that huge amounts of IC production in China will immediately follow. In-country production will increase markedly in 2005 over 2004, however, it will still represents a small portion of the ICs purchased by China. This implies that demand for machine vision equipment used for inspection of semiconductors in China will increase but not dramatically so in 2005.

Since personal computer (PC) sales drive much of the demand for ICs, we also examined available PC sales forecasts.  According to Microsoft, “PC unit growth was very strong in 2004, increasing almost 13%.  However, in 2005, Microsoft believes PC unit shipments will grow 7% to 9%. Worldwide, IDC estimates PC sales growth in 2005 at 10.1%.  SEMICO believes notebook PCs will grow 9.6% in 2005, down from 28% in 2004.

So what do we conclude about the macroeconomic and sector outlook for 2005? We conclude that despite some dark clouds on the horizon, 2005 promises to be a second year of recovery, particularly in North America.  China’s economy will continue to perform strong, although at a somewhat lower level. In Europe and Japan, positive growth is also predicted but at levels below that anticipated for North America.

What does this imply for machine vision sales?  The macroeconomic data we have cited suggests that 2005 is shaping up to be a good year for machine vision sales in North America but less so in some other major regions. (However, it should be noted that GDP growth is not a perfect indicator of machine vision sales.) At the same time, softness in semiconductor sales could dampen demand for machine vision equipment to some extent.  In any case, machine vision equipment sales will probably not hit 2004 levels in terms of annual percent growth.  (For more precise results, please refer to the AIA’s annual Machine Vision Market Study researched by Nello Zuech, available in March 2005.)


The Views of Machine Vision Industry Experts
We next spoke with industry experts, John Merva, President, Tatille USA; Dietmar Ley, CEO, Basler Vision AG; Bob Settle, Director of Marketing, DVT; and Terry Guy, Product Marketing Manager, Eastman Kodak. We asked them about their expectations for the machine vision market as a whole and where the largest opportunities would lie, zeroing in particularly on China.  We also asked them about what major market trends would emerge.  Following are excerpts from interviews that we will post in full on www.machinevisiononline.org at a later date. 
1. How will the Machine Vision (MV) market perform in 2005?

John MervaJohn Merva: “2005 will be a good year for the machine vision market. The machine vision market will benefit from a second, consecutive year of economic recovery. Machine vision products typically involve fairly significant capital expenditures for companies, which must be paid out of their profits. A second year of recovery in the overall economy will mean a continuation of those profits, which will make more monies available for capital expenditures and machine vision products.  At the same time, machine vision products will continue to offer increasing value as a consequence of increasing compute power per dollar.  So machine vision applications will become increasingly affordable and attractive to more and more companies. Taken together, both factors, the continued economic recovery and increasing compute power per dollar, should spur growth in the machine vision market.”

John Stack: “I see steady growth for the MV market in 2005. Industry sectors such as pharmaceuticals, food processing and general manufacturing should continue to experience steady growth along with the economy as a whole. My only worry is the outside possibility John Stack for interest rates to rise quickly and depress overall capital expenditures for equipment. Semiconductors, on the other hand, represent an unknown.  This sector is quite skittish, taking steps forward and then backward.  Growth projections for semiconductors are generally negative, but this might represent over-caution on the part of analysts. Importantly, semiconductor inventories are not particularly high, suggesting that negative growth in this sector is not imminent. So I’m just not sure about semiconductors.”

Dietmar Ley: “From a European perspective I expect the first half of 2005 to exhibit a somewhat weaker performance compared to 2004.  I am hoping for some growth in the second half. The weaker performance in the first half of 2005 in Europe will reflect the weak US dollar. While the weak US dollar will help US machine vision firms to sell abroad, it will present special challenges to European firms, which must find ways of compensating for the weak U.S. dollar. Also, target industries that are very important to the machine vision industry, such as flat panel displays, electronics and semiconductors, will experience to some extent over-capacity conditions, which will soften demand for MV products.”

Bob Settle: “I expect overall growth of around 10% for the machine vision market. For my company, DVT, I expect growth around 25% to 35%. I look at the market from a smart camera perspective in general and from my company’s perspective in particular.  From these perspectives, several factors account for our aggressive forecast:
a) Further development of existing industrial market segments (automotive, pharmaceutical, packaging, etc.) will drive part of the growth. For example, we see growth in the automotive industry in diverse areas (such as the UK, Spain, Brazil and Argentina). The packaging industry will also drive a lot of demand for machine vision products, especially as fully functioning vision systems are offered at lower and lower price points. (Once prices dropped below $2,000 (USD), additional opportunities opened up.)
b) Penetration of new (for DVT) markets made possible by new product introductions. Example: a low cost wafer reader product for the semiconductor market.
c) Growth in expanding geographic markets (Asia in general, China specifically).

Terry Guy: “I think we will have similar performance as 2004, however, I think our industry could see a downturn in 2006 as semiconductor and electronics companies move into a slowdown in 2005. Recent financial news from leading semiconductor manufacturers show chip sales slowing and inventories rising. If not corrected, equipment makers using MV could see a slowdown in 2006 business, as semiconductor and electronics manufacturers reduce their capital equipment budgets for 2006.”

2. Where will the opportunities lie for machine vision companies?

John Merva:  “Opportunities for Machine Vision companies will exist across the board. Importantly, these opportunities will no longer be confined to the factory floor. Opportunities for machine vision products will increasingly exist not just in traditional, manufacturing industries but also new industries.  Important examples of these new industries are traffic management and surveillance.  I am especially excited about the opportunities in non-traditional areas.”

John Stack:  “Geographically, Asia as a whole – not just China, will be an important opportunity for MV companies.  In addition to China, Singapore, Taiwan, Korea and even Japan will represent opportunities.  Taken as a whole, Asia represents the largest manufacturing base in the world. Opportunities will also exist in the States and Europe, as companies increasingly adopt automation on the factory floor to reduce costs. In addition, out-of-the-box machine vision applications such as automotive guidance are starting to become reality. As acceptance grows, this technology and others will quickly integrate machine vision into our everyday lives.”Dietmar Ley

Dietmar Ley: “Regional opportunities for machine vision companies will lie in Taiwan,  China and Korea. There will also be opportunities in some non-manufacturing sectors.  Intelligent Traffic Systems (ITS) will become increasingly important in Asia and eventually also in Europe. The medical sciences will also represent a growing opportunity.  Longer term (within the next five years) I expect nanotechnology to result in important opportunities for the MV industry”

Bob Settle: “Growth will continue in Europe and Asia. Latin America has good potential but may be impacted by local economic and political uncertainties.”

3. Will China represent an opportunity?

John Merva: “There’s a lot of manufacturing moving to China, and that should mean opportunities for machine vision companies. Manufacturing companies in China will increasingly need machine vision products to improve the quality of their products.  (Cost savings will not be an important motive, since labor costs are already very low in comparison to the rest of the world.) At the same time, it will be hard for machine vision companies to realize opportunities in China, unless they understand the local culture.  Longer-term, there is also the risk that the Chinese will take our machine vision technology and make their own machine vision equipment.  So while attractive and real, opportunities in China are challenging in the short-term and will require creative engineering and overall efficiency in the longer-term.”

John Stack: “Definitely, since a lot of semiconductors and electronics production has gone there. But China is not the only game in town in Asia, as I previously mentioned.”

Dietmar Ley: “China (as Taiwan and Korea) will be a very important target market for MV companies. These countries are strongly focused on developing their least cost, mass manufacturing, core competencies.  They will increasingly require the most efficient manufacturing techniques, such as those supported by machine vision, since the margins Bob Settleassociated with mass production are razor thin.”

Bob Settle: “We see China in particular as an opportunity. For that reason we have invested in China for several years by building our infrastructure. We are now established and actively building a distribution network of Automation Solution Providers. Our network of regional offices and in-country application engineers will support this growing customer base.”

Terry Guy: “Certainly, China represents an opportunity for MV equipment suppliers as semiconductor and electronics manufacturers invest in building infrastructure there.”

4. Are there any new, market-related or technological developments that we should look for in 2005?

John Merva: “I believe cameras with IEEE 1394 and GigE connections will be very hot in 2005.  Dedicated processors will also pack more bang for the buck.”

John Stack: “I am rather discouraged regarding technological developments in the traditional MV industry. In contrast to other industries, I do not see the long-term, strategic evolution happening in our industry.  A great many non-traditional machine vision applications have sprouted up over the past few years. In some cases they have begun to catch the attention of the global players such as Microsoft, Intel, NEC, HP, Automotive Companies and a host of others. This makes the chances for a disruptive technology being developed highly probable.  Although most machine vision companies are in no position to compete with such giants, they do have one major advantage: implementation expertise. This makes the traditional machine vision community the perfect vehicle, right down to the one-man system integrator, for developing ‘stand alone products’ that implement state-of-the-art vision technologies, even if that technology comes from outside the industry. The question in my mind is whether or not traditional machine vision companies will venture off the factory floor to take advantage of the opportunities that await them.”

Dietmar Ley: “Smart cameras will become increasingly important as an architecture for low-end applications. Therefore, they will cut into sales for low-end applications based on a PC architecture. Low-end frame grabbers, for example, will become vulnerable.  For that reason, a producer of low-end frame grabbers must make the transition to a solutions provider. Demand for higher-end frame grabbers will not be affected to the same extent. Going forward, MV products must become easier to operate for lower skilled people. Products must also comply with new governmental regulations such as RoHS (Restriction of Hazardous Substances), which is going into effect in Japan at the end of 1st Quarter 2006 and in Europe.  MV companies seeking to do business in these important markets should review their production processes to verify that their products do not contain harmful substances such as lead.”

Bob Settle: “In the near term, technology may not look much different than it does now. Terry GuyThere will be advancements in lighting and image acquisition sensitivity, but probably not major paradigm shifts. However, the role and extent of customer support will change. Customers will expect higher and higher levels of support. MV Companies that can deliver on those expectations will be successful. Ease-of-use will be important to many users (especially lower-end ones) and application-specific solutions that require little – or no, setup will increase in importance. Availability of comprehensive, low cost (or free) training will continue to be a competitive advantage for some MV companies.  Also, software, which  currently provides a major revenue stream for many MV companies, will become less expensive.  Many customers feel that they have been ripped off with software fees and upgrade charges, and will apply pressure to reduce this large element in their total cost of ownership of MV systems. “

Terry Guy: “I think we are just beginning to see what smart cameras can offer. Currently, they can perform inspection tasks such as gauging, 2-D code reading, and presence/absence to mention a few. In the future I think we will see many more “inspection” tasks being performed that ultimately will drive increased yield and quality of manufactured products.”

















The author invites questions about this article.
Please send your feedback to: content@machinevisiononline.org




Paul Kellett





There are currently no comments for this article.

Leave a Comment:

All fields are required, but only your name and comment will be visible (email addresses are kept confidential). Comments are moderated and will not appear immediately. Please no link dropping, no keywords or domains as names; do not spam, and please do not advertise.

First Name: *
Last Name: *
Your Email: *
Your Comment:
Please check the box below and respond as instructed.

Search AIA:

Browse by Products:

Browse by Company Type: