Producing the next generation of computer chips requires as much attention on an industrial scale as cooking a gourmet meal. The finest materials, techniques, tools and of course the sharpest minds must be used to create what is transformative. If even one of him is missing from the kitchen, the meal is inadequate.
In this regard, the Department of Commerce will soon commit $11 billion to research and development under the CHIPS Act to revive America’s sluggish chip manufacturing industry, which produces only 12 percent of the world’s chips. I plan to lead the banquet. In passing CHIPS, America asserted a bold desire to return to the forefront of chip manufacturing. But there is a big gap between desire and action. not easily surpassed.
CHIPS Act funding is a lot of money, but the numbers are no guarantee of success. Chip manufacturing is an almost incomprehensibly precise, difficult and expensive business. Ensuring that the U.S. sits at the table with the world’s biggest manufacturers will require innovation and collaboration on a grand scale.
Lowering the barriers to participation and funding – using the finest materials and the finest equipment and inviting the most astute talent – is critical to the future of American CHIPS-inspired semiconductors. The nearly $600 billion global semiconductor chip industry produces over a trillion chips each year, which are used in everything from cars to coffee makers. A “more is more fun” approach will create a vibrant, fast-moving network of innovations that create the breakthroughs needed for success.
So far, much of the focus has instead been on the industry’s biggest, and often the slowest, companies. America thrives on innovation, but many of the most creative minds in small businesses and universities suffer from the field’s notorious R&D costs and access to expensive tools and facilities (often called “fabs”). Locked out by lack. Required for prototyping. As any fine chef will tell you, if you don’t have a kitchen to cook in, you can’t create anything innovative.
There are two ways the United States allocates these R&D funds. First, relying on the “famous chef” approach and potentially placing big bets on a few names in hopes of a magical breakthrough, this is a risky proposition.Or you can build more kitchens and invite them more The chefs start cooking, making many small bets on the most innovative minds, creating an innovative chip-making co-restaurant line that makes steak, not sizzling steak.
First, you will need more kitchens. It is necessary to build a joint experiment facility where researchers from industry and academia collaborate. This will foster a hands-on research community that brings together leading engineers around a common purpose and opens doors to innovators across the country. These technology centers range from hardened university factories run by experienced engineers to industry-scale facilities that showcase new ideas of product value at scale.
Next, we need to build a “digital twin” computational model that can emulate the entire manufacturing process, including the details of the process tools and process conditions. Digital twins enable researchers to quickly evaluate options and accelerate process and device technology discovery. By collecting massive amounts of data, AI models can help humans fine-tune delicate manufacturing processes and automatically detect anomalies during manufacturing to improve production yield and quality. . Unthinkable today, such a digital model would be a ticket to America’s leadership in semiconductor technology and manufacturing that would rely less on humans looking at spreadsheets to make decisions.
Implicit in this vision is the critical role of data sharing across the community. Open access facilities are networked nationally and freely distribute data throughout the community while protecting confidential information. The impact of this digital twin will grow exponentially as the research community expands and new knowledge drives innovation.
Ultimately, such a network will reduce R&D costs for all players, reduce risk, and bring innovative ideas to market faster. A shared prototyping environment can drive the production of all kinds of chips, including logic, memory, storage, and specialized technology, in the same way that a well-equipped kitchen can produce a wide variety of dishes overnight. .
But all this talk about improving access overlooks another key ingredient to success: talent. Good access means nothing if you don’t have an exceptionally skilled chef who wants to step into your kitchen. We must foster a healthy industry with healthy profit margins that supports career growth, attractive working conditions and a good work-life balance to attract the next generation of engineers. Such talent does not grow on trees. It takes 10 years or more for a high school graduate to complete a PhD. If you want to be one of the industry leaders 10, 20, 30, 100 years from now, you must recruit and train new talent today.
The goals of the CHIPS Act are high. Regaining global leadership in semiconductor manufacturing will not be easy, but it is within reach for a capable and ambitious country like the United States. The recipe is even simpler than this. It’s about training the best people. Create an open and collaborative ecosystem. Get the most advanced tools.To give permission everyone Join and share data. Then enjoy a delicious 4-course meal from lab to factory. thank you.
This is an opinion and analysis article and the views expressed by the author are not necessarily those of the author. Scientific American.
