A key question in the commercialization of small (and large) products is: should you buy it or make it yourself? This question is related to a few other questions about the company strategy, its technology, and its environment: Do we have the needed background? Can we afford to invest in new equipment? Is the work so confidential that it has to be done in house? Can we exploit available existing/standard processes? Or is the technology changing too fast for the infrastructure that is available externally?
In the early days of semiconductors, every company had its own wafer-processing infrastructure, they packaged and assembled their own products, and they often even developed their own specialized equipment. However, the buying of equipment and the outsourcing of back-end processing (packaging and assembly) soon became standard practice. Other parts of the business chain followed, and slowly front-end semiconductor processing also was transferred to external foundries.
|
Fabless and fablight IC companies are common now. The supply chain has grown very complex, with lots of niches filled by specialized high-tech companies. Products now differentiate by design, not by technology. Initially, with MEMS, a similar approach was quite common: easily accessible university equipment supplemented with affordable 100-mm equipment provided easy access to production. Later, however, when the processes became more standardized and infrastructure became more expensive as the industry went to 150/200 mm wafers, the first MEMS foundries saw the light of day. Now some foundries are amongst the most successful of MEMS companies.
Still many fabless companies are using university and institute labs’ equipment instead of commercial foundries. The Fraunhofer Institutes in Germany are pretty successful in offering foundry services, e.g. Fhg-IMS Duisburg is producing pressure sensors for Elmos, and the equipment of FhG-ISiT Itzehoe is used by Plan Optik. This proves that outsourcing to universities and institutes is an option, even for the production of large quantities of MEMS products for demanding markets.
Another important aspect that needs mentioning is the increasing trend towards IC-compatible MEMS technologies. Newly developed sensors, especially in high-volume applications, are increasingly designed into an IC factory, profiting from its high standards and economy of scale. Commercial foundries are driven by the market, and they will make (nearly) any attempt to follow their customers by extending their range of technologies.
When considering materials other than silicon, the gap between the OEMs, component suppliers, and the existing infrastructure is wide: a substantial number of users need polymer/plastic parts with small dimensions. These can neither be delivered by the traditional moulding companies nor by the silicon foundries. Some European foundries found a solution to that problem by aligning themselves with expert groups or projects funded by the European Commission; e.g. the foundry Tronics Microsystems in France did so by participating in the Microbuilder project and Silex did so by becoming a partner in the INTEGRAMplus project, both running under the Europractice label. Quite often the gap is being filled by expert companies such as Micronit (glass processing) and Epigem (polymer parts). The Europractice family of projects, initially structured into separate competence centers, design houses, and foundries, nowadays acts as an integrated design and manufacturing service provider that addresses even quite complex manufacturing challenges.
Most of the successful MEMS start-ups of the last few years are fabless: Knowles, Discera, Akustica, SiTimes, Virtus, etc. But they are not launched by people starting from scratch; their founders are often serial entrepreneurs with significant track records in the industry. Can you do this also as a start-up with limited experience in business? The answer is yes, assuming you are part of a network of specialist companies. An example of such a cluster is the Twente cluster in the Netherlands. The students are so indoctrinated in the skills of entrepreneurship, that when asked what their preferable career would be after leaving university, one in seven answers: starting up a company—quite unusual for Europe! This has led to many start ups in the last 25 years, and the numbers are rising fast, now at about 10 per year.
|
|
Patric Salomon, Germany, offers services with a focus on marketing, public relations, and business strategy in micro- and nanotechnology. He is the vice-chair of NEXUS, the European Microsystems Association.
Henne Van Heeren, The Netherlands, is a specialist in MNT manufacturing, packaging, and technology in business-related issues. He is a member of the NEXUS Task Force on market research and an author of the MANCEF MNT roadmap.
Salomon and van Heeren advise industry and public bodies on MNT product commercialization and business development. They publish a report series called enablingMNT Industry Reviews, www.enablingMNT.com.
Published in R & D magazine: Vol. 50, No. 2, April, 2008, p.40.
