Copresenter: John Solis, Senior Software Manager, Tokyo Electron US
Digital twins can be constructed at various hierarchical levels of industrial systems. Often, completely different models are created at these different levels, resulting in limited interoperability due to disparities in scale and data requirements of these models. In this presentation, we explore how building data models for semiconductor capital equipment that are consistent with best practices from the International Society of Automation's ANSI/ISA-95 standard allow for scalability and adaptability to new use cases. Digital twins of etch chambers originally built to enable predictive maintenance can be leveraged across a fab to ramp up new tools more efficiently. Combined with select but relevant fab data, the models running these digital twins can improve productivity of existing equipment fleets while reducing time to ramp new capacity. They can also be scaled up and adapted to different scenarios to ensure robust supply agility in the event of perturbations in demand. Needs of new fabs plus robustness against unexpected disruptions can be more accurately factored into supply plans.
