For a technology that is an offshoot of the semiconductor sector, micro-electromechanical systems have remained surprisingly immune to the boom/bust cycle that semiconductors repeatedly experience. While that does appear to be changing ever so slightly, the MEMS sector continues its upward trek in terms of revenue, despite tough economic conditions.

Perhaps the greatest issue facing one sector of the MEMS industry is fab overcapacity, a situation that appears as if it will only worsen in the next year.

Several state-of-the-art MEMS fabs are being built and are expected to be online by the end of 2003, and more traditional semiconductor foundries continue to enter the MEMS fabrication business.

That puts considerable pressure on the established MEMS fabs to maintain a break-even level of capacity. As a result, this is a somewhat risky time for fabless MEMS companies to choose a production partner; many already have contingency plans for secondary, and even tertiary, sources.

One of the biggest stories over the past two years has been the outpouring of venture capital into the MEMS industry, to an historic level of more than $1.2 billion in 2000. For the first six months of 2002, startup equity was only one-sixth of that. However, while VC funding is certainly down, it is by no means out.

That's an important point for an industry that by and large has been built on funding from the Defense Advanced Research Projects Agency and other government bodies, not venture capital. More VC firms than ever before are aware of MEMS technology and, despite tightened purse strings, still see considerable opportunity there.

Meanwhile, device innovation and commercialization efforts continue, and every major market has embraced MEMS technology. Mature applications are still experiencing growth, and new applications continue to be found.

There is a clear migration path of device commercialization. MEMS devices that have found market success to date include accelerometers, pressure sensors, nozzles, the lab-on-a-chip and gyros. Improved sensitivity and lower prices are allowing these devices to move into more new markets and applications than ever before. In fact, sales of this particular group of "first wave" devices are the backbone of the MEMS industry and will remain so for quite some time.

We're now seeing the second wave of devices, which companies have been striving to bring to market over the past several years. They include mirror arrays and a much more diverse range of sensor technologies, such as biological, chemical/gas, infrared, humidity and biometric sensors. Most have already entered the market to some extent, but their full potential has not come close to being realized.

As these technologies mature and prices decline, they will expand quickly into many more applications beyond the initial focus of the companies developing them. The wireless networking of the sensors in this category will be especially important.

The next wave of MEMS device commercialization-involving products that, for the most part, are still being developed-is definitely emerging. Sufficient numbers of companies are bringing these devices to market to fuel expectations of strong revenue growth over the next few years. So far, this category includes microphones, RF MEMS and, to some extent, pumps (particularly implantable ones) and probe arrays. The latter are being investigated for applications ranging from transdermal drug delivery to data storage. Beyond these four device types, MEMS-based power solutions are being worked on, but it's still too soon to determine their impact.

With such expanded device innovation, worldwide MEMS revenue is forecast to grow from $3.9 billion in 2001 to $9.6 billion in 2006. The growth will primarily be driven by applications for the communications and consumer markets, despite the severe retrenchment being experienced in the optical-networking sector. While the short-term opportunity for MEMS in optical networking has all but disappeared, the long-term outlook remains good-depending on how well MEMS companies involved in this sector are able to weather the downturn. Commercialization of RF MEMS for cell phones is just beginning.

All the remaining market segments will see somewhat lower growth rates but will be steady. The automotive, computer and industrial markets have been, and will continue to be, the bread and butter of the MEMS industry. And despite those markets' maturity, more applications continue to be found.

The final frontier, so to speak, appears to be the medical market, where some of the most innovative MEMS devices are being developed and commercialized. And while the development cycle here tends to be longer than it is for other markets, and there are hurdles-the FDA, for one-with which other sectors need not be concerned, the use of MEMS in medicine is garnering attention, which translates to revenues.

Despite some growing pains and pressure from current economic conditions, the MEMS industry appears well-situated for continued upward growth. The bottom line is that the technology is doing what it does best, and more so than ever before: enabling new applications and disrupting existing ones.

Marlene Bourne is Senior Analyst, MEMs, at research firm In-stat/MDR (Scottsdale, Ariz.).