Developing a consensus-based plan for the next wafer size transition
01/01/2007
As part of the plan for a next wafer size transition, International Sematech Manufacturing Initiative (ISMI) is developing a strategy for steering design evolution in 300mm-as a precursor to scaling up to 450mm-including the timing and risks associated with design changes needed for next wafer size manufacturing insertion. Industry consensus on the direction of key aspects of the next-generation factory will be essential in minimizing R&D costs. This article describes ISMI’s collaboration with the industry to develop a consensus-based approach to these issues.
The topic of the next wafer size evolution is heating up in the semiconductor industry. As part of the plan for this transition, a strategy of steering continuous design improvements in 300mm-as a precursor to scaling up to 450mm-is being proposed. There are also questions over the timing and risks associated with the design changes needed for next wafer size manufacturing insertion. Industry consensus on the design direction of key aspects of the next-generation factory will be essential in minimizing R&D costs for different designs addressing common tasks. Industry consensus also is needed to keep various design areas synchronized and prioritized. This article shows how device makers are collaborating with the industry to develop a consensus-based approach to these issues. It also identifies challenges to achieving success in collaborative industry efforts.
ISMI’s 450mm program
ISMI, a consortium of 15 device makers, provides productivity solutions for current and future challenges in manufacturing fabs to help the membership achieve best-in-class productivity levels. The objectives are accomplished by providing platforms for collaboration among members and development activity in key areas identified by members. Within ISMI, there are programs in factory productivity, equipment productivity, environment, safety & health, metrology, and 450mm.
A reduction in the rate and pace of productivity improvement that the industry has experienced before 2006 is of particular concern to the membership. The ISMI economic model has characterized this effect and predicts a slowing of productivity improvement, mostly due to a slowed pace of technology design shrinks and innovations in design layout (Fig. 1).
 Figure 1. Plot illustrating the slowed pace of technology design shrinks and innovations in design layout. |
At the end of 2005, an ISMI 450mm program was established to focus on improving the industry productivity trend. The program employs a concept called 300mm Prime, which couples 300mm design improvements that can be scaled up in a 450mm factory environment. The intent is to improve 300mm productivity in a way that minimizes the need for revamped or revolutionary design in 450mm. A key lesson from previous wafer size conversions is that staging new concept implementations will lower risk, minimize design cycles, reduce design development costs, and achieve earlier overall success.
Historically, wafer sizes have changed every 10 to 12 years, the last being 300mm in 2001. The International Technology Roadmap for Semiconductors (ITRS) projects the next wafer size change in 2012. When the industry is indeed ready for this transition, 300mm Prime presents an evolutionary path to 450mm. As such, much of the 450mm program at ISMI is geared to give direction and definition to 300mm Prime.
The ISMI members have established focus teams comprised of experts in starting materials, factory architecture, factory simulation, and economic modeling. Together they are collaborating on a common vision for the next-generation factory founded on 300mm Prime principles surrounding scalability to 450mm. One goal is to be able to use 300mm Prime designs in existing factories to provide a continuum of productivity improvements from 300mm to 450mm.
The ISMI program is exploring options related to the next wafer size transition. This exploration includes the development of a next-generation factory vision, factory simulation, economic modeling, and starting material assessments. The creation of an integrated 300mm and 450mm vision will enable a path from the 300mm fabs of today, known as 300 Classic, to a next-generation 300mm factory, known as 300 Prime, to 450mm. Minimizing R&D costs, accelerating learning at lower costs, lowering risks, and enabling improvements at 300mm were all factors in the decision to create such a long-range plan.
The transition from 200mm to 300mm wafers was analyzed to learn what was done well and what could have been done better. Increased involvement of the supplier community in the development of the transition plans was highlighted as a key area for improvement in future endeavors. Significant industry communication and coordination have been planned into the ISMI program in the spirit of continuous improvement.
Visions and roadmaps
Industry initiatives on the scale of a wafer size transition require global coordination and consensus. A high level vision provides direction for such change and innovation. A vision should be supported by analysis and consensus on objectives such as productivity targets, cost and benefit goals, and technology advances. Figure 2 illustrates the general flow of the life cycle of requirements, from initial vision through delivered implementations. This article addresses the need for collaborative efforts up to the actual implementation of solutions, taking into account business and economic considerations, the definition of technical requirements, and standards development.
 Figure 2. Sequence of requirements for the wafer size transition. |
Roadmaps, such as the ITRS, provide a 15-year outlook that includes technology node insertions, wafer transition timing, and factory productivity targets. The initial, fundamental decisions and common semiconductor manufacturing requirements require identification, analysis, and prioritization of issues that provide potential for productivity improvement.
The ISMI Factory Architecture team is working to provide a next-generation factory vision and manufacturing operations assumptions, thereby achieving an historical cost reduction in processed silicon and cycle time reduction targets. This vision will also drive an industry standards roadmap. Furthermore, supported by ISMI factory simulation data, the factory architecture could drive key decisions on wafer carriers, automated material handling systems (AMHS), automated reticle handling systems, production equipment, and related factory attributes. This focus team will create a consistent factory architecture vision for 300mm Prime and 450mm, based on the best known alternatives, which will accommodate multiple business models.
The Factory Automation Standards Tracking (FAST) Roadmap, sponsored by ISMI and SEMI, surveys and reports device makers’ intentions to require new and emerging software standards and timing for delivered conformance. The FAST survey also tracks supplier status and plans, enabling better planning for design, development, and implementation. A roadmap such as FAST should be part of an initial standards strategy and refreshed as the standards evolve to take into account changes in industry timing. This survey requires consistent and representative participation by device makers and suppliers to provide useful, meaningful information.
Economic analysis
Consensus on an industry strategy for transition must be driven by rigorous economic analyses of the markets, technology, and cost dynamics. The ISMI industry economic model (IEM) (Fig. 3) tracks industry productivity, enables comprehensive analysis, and can be used as a platform for a quantitative dialogue with the supplier community about the pace, adoption rate, and cost associated with technology and manufacturing innovations.
 Figure 3. ISMI industry economic model. |
The IEM is a one-of-a-kind tool that integrates, at an industry level, semiconductor process technology, wafer diameter, and factory and equipment configurations along with many core strategic manufacturing and development planning functions. The logistics, algorithms, and assumptions used in the IEM have been validated and can generate scenarios that not only examine the drivers of past, present, and future trends in productivity, but also assess the impacts of demand fluctuation and business cycles.
In the ISMI/SEMI joint productivity working group (JPWG), device makers and suppliers are working to forge a fundamental understanding of the magnitude and timing of the R&D investments required to maintain the historical productivity improvements in semiconductor manufacturing. A joint economic analysis team (JEAT) was subsequently chartered to identify the root causes of the productivity slow-down and to establish a viable methodology to evaluate proposed initiatives for optimizing productivity.
A broad-based, highly interactive user group is key to realizing relevant results. The IEM business logic is calibrated through historical validation, but metrics must be continuously improved through information sharing and cooperation between device makers and suppliers. Further collaborative studies are required to validate the assumptions to create an unbiased resource for the industry. ISMI has made assumptions for 450mm wafers in the economic model, along with identification of technical and cost challenges. Collaborative efforts between device makers and silicon suppliers will be needed to ensure that the analysis of economic and technical feasibility and timing leads to an informed industry strategy for 450mm wafer readiness.
Factory simulation
A 300 Prime analysis-a Pareto of potential productivity improvement options-is being evaluated to target a 30% cost reduction and a 50% cycle time reduction by 2012. Factory simulation is underway to assess the productivity opportunities of potential solutions.
 Figure 4. Core components of factory vision. |
The ISMI factory simulation relies on the factory architecture team to help define simulation input parameters. The process will be highly iterative. As simulations are built and analyzed, results are provided for factory architecture development. The intent of the ISMI factory simulation is to simulate chosen variables for factors that most affect factory operations. Initial models will look at the effects of carrier capacity and the increased moves that are imposed on a material handling system. ISMI membership is characterized by a broad range of business models, and the simulations are expected to be representative by simulating added factors including factory capacity, product mix, and issues affecting equipment performance.
Guidelines
Once the factory vision is understood, more detailed guidance will be provided in ISMI factory guidelines. To achieve cycle time and cost reduction targets, device makers may require new AMHS transportation systems, AMHS storage stockers, carrier design, automated reticle handling systems, and production equipment guidelines. The ISMI Factory Architecture team has identified key factory core components, as shown in Fig. 4.
 Figure 5. Potential standards change areas. |
The factory guidelines will create a platform for effective discussion between device makers and suppliers about factory requirements. Moreover, the factory architecture will drive industry standards for these key core components. Potential changes to the standards are listed in Fig. 5.
Standards
New initiatives should comprehend the standards needed to enable significant changes for future productivity improvements. ISMI members will share progress as they develop a factory vision, implementation options, a standards roadmap, and user guidelines in the newly formed SEMI Manufacturing Technology Forum (MTF) to identify common requirements and solutions that require standards. Participation by organized device maker and supplier groups is essential to achieving an industry-supported vision and roadmap.
Standards development in SEMI is actively supported by participation from ISMI and its member companies working with supplier participants. A challenge in standards development is maintaining consistent, representative participation by device makers and suppliers. Collaboration in the MTF to agree on the standards strategy and roadmap may improve the support for and participation in new standards initiatives.
ISMI also participates in SEMI initiatives to seek ways to accelerate and improve standards development through early standards prototyping and validation. ISMI and suppliers have collaborated on some early standards prototyping and validation to improve the definition, implementation, and interoperability of standards such as recipe and parameter management (RaP) and equipment data acquisition (EDA). Scenarios and use cases provide application and implementation guidance to improve conformance to actual operational requirements. Without these approaches as mainstream practices, the resulting standards can be tested only in real production environments with several cycles of costly revision. Moving forward, emphasis is expected to be placed on prototyping and early standards learning to improve standards development and implementation processes.
Conclusion
Many of the approaches discussed in this article are being piloted or executed at a limited level, particularly early prototyping and standards validation in the software standards arena. The industry needs to find ways to adopt these approaches as mainstream best practices to achieve their full benefit throughout the life cycle of requirements.
A key to the consolidation of issues and identification of viable, cost-effective solutions is the representation of industry groups to bring proposals to the discussion table. ISMI represents a consortium of device makers who work together to develop consensus positions and direction. SEMI is sponsoring the formation of an Equipment Suppliers Group (ESG) to provide a forum for suppliers to discuss and prepare organized input to the process. The efficacy of the MTF will partially depend on participation of organized supplier groups such as the ESG.
Major decisions for future investment will be made based on the analysis, data, and information that is gathered and evaluated through industry dialogue and collaborative initiatives. The decisions made will be as good (or as bad) as the knowledge and understanding gained. SEMI’s MTF and ESG will require organized, representative participation. Collaborative efforts among suppliers and device makers for significant industry initiatives are essential to realize cost-effective productivity improvement.
Acknowledgments
The authors would like to acknowledge the contributions of ISMI assignees Thomas Abell (Intel), Kun Tsang Kuo (TSMC), and Edward Bass (Intel), and ISMI’s Robert Wright. Advanced Materials Research Center, AMRC, International Sematech Manufacturing Initiative, and ISMI are servicemarks of Sematech Inc. Sematech, the Sematech logo, Advanced Technology Development Facility, ATDF, and the ATDF logo are registered servicemarks of Sematech Inc. All other servicemarks and trademarks are the property of their respective owners.
Reference
1. ITRS FI Table 89a Factory Integration and Control Systems Technology Requirements: Near-year terms; http://public.itrs.net.
Jackie Ferrell received her bachelors in business administration from the U. of Texas in Austin. She co-chairs the ISMI 450mm starting materials focus team, chairs the SEMI NA Regional Standards Committee, and is a voting member of the US Technical Advisory Group to ISO TC209. Sematech, 2706 Montopolis Drive, Austin, TX 78741-6499; e-mail [email protected].
Denis Fandel received his BS in mathematics from the U. of Wisconsin and is manager of the Industry Economic Modeling (IEM) project at International Sematech Manufacturing Initiative (ISMI). Earlier this year, he accepted an assignment in the new ISMI 450mm Program to manage the Transition Strategy and Assessment project.
Joseph Draina received his BS in mechanical engineering from Rutgers U. and his masters in mechanical engineering from Columbia U. He is currently on assignment from IBM, where he is a senior engineering manager. In April 2005, Draina was named the associate director of International Sematech Manufacturing Initiative (ISMI).