Data is the Game Changer for Manufacturing Excellence
January 19, 2022
With more active devices today than ever, data is growing at exponential rates and systems are becoming more sophisticated and ever-more complex.
Behind the creation of these systems and devices are roughly 200 process steps within a semiconductor manufacturing process or flow, which increasingly require fine tuning of materials to support current and next generation technologies.
Over the past few years, the dimensions for each technology node have tightened. As new materials have been introduced to meet performance targets for new devices, new challenges are rising around materials interactions with the manufacturing processes. Each individual process has become more and more sensitive to any variation. That is not only from the macro properties of the material, but more commonly now, also with sub-suppliers process or some variation that led to an impurity.
The semiconductor industry, including material suppliers as well as device manufacturers, have recognized the need to evolve the approach on the quality and analytics for materials. A new industry-wide data collaboration and analytics platform is needed such as one enabled by Athinia, through a partnership between Merck KGaA, Darmstadt, Germany and Palantir Technologies.
Providing advanced data analytics can help limit the costly impact of quality or performance excursions across the value chain, from supplier to semiconductor fabrication plants. It will also help fabs manage faster innovation in manufacturing processes in a single, secure platform that will support improved incoming material quality and increase supplier engagement. Suppliers benefit from internal efficiency gains through smart data integration and can be a better partner for the fabs they serve.
The understanding of what data, how to bring the data together, and then how to do it in a way that's secure so that you drive the insights without having concern of IP contamination is key.
By leveraging artificial intelligence and big data analytics in a secure and enabling data sharing platform, material suppliers and device manufacturers can share big sets of data around materials and fab processes to enable the next advancements in materials quality as well as the supply chain. Off-the-shelf machine learning models are used to analyze correlations in the data through the Palantir Foundry platform.
A collaborative data sharing platform can even help with challenges such as the chip shortage by enabling high utilization efficiencies of device manufacturing plants by minimizing or eliminating any material disruptions. This also helps any given material supplier manage the supply chain more effectively and make sure that they can continue to have a sustainable supply, further minimizing any impact of materials to the chip shortage.
Using this data driven decision making starts with a solid data foundation where security is paramount. Customers maintain sole ownership of their data, retaining control in a secure environment where their IP is preserved. Athinia offers capabilities to anonymize (remove meta data, i.e. column names) and scale to allow for a secure flow of information and reliable statistics.
The industry benefits of a platform like this are wide ranging, including:
- Improving the time to market for new materials for next-generation technology nodes,
- Enabling higher yields,
- Driving to zero defects for device manufacturing,
- and improving supply chain delays.
All of which will help accelerate the delivery of new consumer electronics.
Recently, representatives from Athinia, Merck KGaA, Darmstadt, Germany, Palantir Technologies and SEMI Americas discussed this topic in a virtual CES 2022 panel discussion.
You can view the replay here.