Bill McClean, IC Insights, Inc.

ABSTRACT

After steep declines in 2001 and 2002, worldwide semiconductor industry capital spending is forecast to show a 15% increase in 2003 to $31.9 billion. Total semiconductor industry capital spending in 2007 is expected to reach about $44 billion.When using the $44 billion figure for 2007, the 1994-2007 CAGR for semiconductor capital spending is only 5.2% (about 7 points less than the 12.1% 1978-2002 CAGR). It is interesting to note that the 1994-2007 CAGR in the semiconductor market is forecast to be only 6%, illustrating the tight correlation between semiconductor market growth and capital spending. 

Thomas R. Pomorski, Brooks Automation

ABSTRACT 

The one constant in manufacturing operations is change. Learning and improving faster than the competition may be the only lasting competitive advantage in this new environment. This presentation will review continuous improvement practices and methodologies, such as Total Productive Manufacturing (TPM), Lean Manufacturing and Learning Organizations that are commonly deployed to optimise manufacturing effectiveness. Discussion will include the role of automation in continuous improvement programs, as well as the contribution of equipment productivity to overall manufacturing performance during each stage of the fab life cycle.

Robert C. Reuel, Intel, Rio Rancho, New Mexico, USA

ABSTRACT

The incorporation of lithographic APC solutions at Intel has shown how a simple concept such as active feedback control can be rife with obstacles and surprises. These encounters included automation and implementation challenges, myths, manufacturing learnings, and shifts in company policy. As the APC system began to mature and approach production proliferation, misconceptions and myths arose. Some engineers expected to see large capability improvements across all layers or to be able to associate end of line results to APC implementation. The fact that APC was implemented did not mean that process monitoring could be neglected or tool pairings could be disregarded. Processes could no longer be intentionally run off target. Metrology was found to be more important than ever, with the potential of having inaccurate measurements snowballing into multi-lot rework excursions.

JOHN W. FOWLER, W. MATTHEW CARLYLE, GEORGE C. RUNGER & ESMA S. GEL, Arizona State University, USA
SCOTT J. MASON, University of Arkansas, Fayetteville, AR, USA
OLIVER ROSE, University of Würzburg, Germany

According to the 1997 National Technology Roadmap for Semiconductors (NTRS), the cost of equipment is approaching 90% of wafer fab capital costs. The NTRS indicates that in order to utilise this equipment effectively, "significant improvements in factory planning/scheduling" are required. However, scheduling semiconductor-manufacturing facilities is a very difficult problem and is among the most complex scheduling problems encountered today.

RICHARD WANG, SHELBY LAURENTS & CHARLES LYNN, Fluor Microelectronics

ABSTRACT

Along with traditional engineering, procurement and construction (EPC), the profitable nextgeneration 300mm fab will require the integration of physical, logical and people systems on an enterprise-wide basis. This article looks at the inherent challenges within the fab itself, along with outside-the-industry models that are now being refined to suit semiconductor manufacturers.

ABSTRACT

The semiconductor manufacturing industry faces a never ending pressure to improve productivity and process flows. This is naturally accompanied by the drive to eliminate unnecessary tasks and their associated costs, by reducing efforts in manufacturing and service functions. The methodology described in this article focuses on a team based approach, where people from all parts of the organisation are encouraged to be creative and to contribute towards the utilisation of the intrinsic improvement potential.

What will 12 inch semiconductor fabs look like? There has been so much effort to establish standards and guidelines for 12 inch semiconductor fabs that we may feel that the era of 12 inch fabs has already come, and wait for high-volume production. But there still exists a question mark. This article covers several topics, ranging from defining the initial layout design concept to analysing the layout from the viewpoint of material flow.

The used and surplus equipment business is now a multi-billion dollar global market place. The trade cycles of the Semiconductor industry are windows of opportunity for the astute entrepreneur. The evolution of secondhand and surplus industry is discussed and guide lines for successful future business operations are indicated.

New financial tools are offered to the semiconductor manufacturers and can help maximise the lifecycle profitability of semiconductor equipment. But everybody's business situation is unique; does your leasing company offer you an equipment control and ownership structure unique to your business needs and situation? In this paper we will take a look at different kinds of leasing options, important facts you need to evaluate before entering a leasing deal and why "flexibility" is the key issue to look out for when coming to an agreement.

JOCHEN HILDEBRANDT & CLAUS-PETER DEISSLER, 4stoXX AG, Stuttgart, Germany

ABSTRACT

Among other things, the semiconductor industry is typified by the increasing difficulty with which knowledge is transferred. Knowledge sources and sinks are widely separated, and transfers are both highly filtered and, usually, uni-directional, from supplier to customer. Which is more than unfortunate, because it is knowledge flow in the opposite direction that presents a decisive opportunity for the industry: in practice, nobody knows better than the user where the problems are and which improvements could make the biggest differences.

LEON F. MCGINNIS, Georgia Institute of Technology, Atlanta, GA, USA, JOHN FOWLER, Arizona State University, USA, STANLEY GERSHWIN, Massachusetts Institute of Technology, USA, ROBERT LEACHMAN, University of California, Berkeley, USA
JIM IRWIN, IC Irwin Consulting,USA, DOUGLAS SCOTT & MITCHELL WEISS, PRI Automation, Billerica, MA, USA

ABSTRACT

The design of Fabrication plants for future high resolution products is a vital and daunting task. The distant vision of a totally automated factory with minimal service staffing comes to mind as a designer's dream. However, reality demands that the complexities of design require very careful thought and engineering expertise when investing the enormous capital outlay that a modern plant requires.

ABSTRACT

Factors involved in successful e-strategy for the semiconductor industry are outlined in this paper. Peer-to-peer networking applies the Internet both internally and externally to customers, suppliers, partners, etc. These collaborative workspaces can be applied throughout the technology lifecycle and examples of this are given.

ABSTRACT

The fabless-foundry business model promotes faster time-to-market for new products with relatively low business risks to all parties. Factory and tool automation is a critical success factor of the "fabless-foundry" business model.

ABSTRACT

Companies that manufacture chips are pressured to run at peak efficiency. Reducing down time by 1% on the 50 most critical tools in a typical factory could save the fab one hundred million dollars annually. The solution involves a trade of data for knowledge between fabs and equipment suppliers. The most important requirement is a secure, global data network.

Controlling inventory is a constant headache for Fab plant managers. The amount of capital consumed in this operation can be critical to the success of fabrication profitability. Increasingly, outsourcing this burden along with equipment support and maintenance becomes an attractive proposition for plant management.

ABSTRACT

Pure-play foundry companies (companies that focus entirely on producing ICs for other companies) are some of the most aggressive spenders on IC capacity additions. Their sales are expected to approach $10 billion in 2000. Growth in foundry business can be expected to attract new competitors. Some new entrants may be in the form of joint ventures. IC Insights believes that the emergence of pure-play foundries will have a major impact in the future IC industry.

ABSTRACT

Implementing "islands of software" will address the challenges of the multi-platform environment that exists in fabs worldwide. The main issue is being able to use the latest software tools and methods while working with the legacy systems and other systems still in use throughout the industry.