Engineering Performance Optimization

As advanced technology continues to evolve and integrated circuit (IC) geometry keeps shrinking, the need for tighter manufacturing process and quality control becomes extremely challenging. TSMC has tailored its manufacturing infrastructure to handle a diversified product portfolio that uses strict process control to might tightened specs, and higher product quality, performance and reliability requirements. TSMC’s process control systems are integrated with numerous intelligent functions to achieve excellence in both quality and manufacturing. Through intelligent detection, smart diagnosis, and cognitive action, the Company produce remarkable yield enhancement, quality assurance, workflow improvement, fault detection, cost reduction results, while shortening its R&D cycle.

To meet 5G’s stricter quality requirements for mobile, high- performance computing (HPC), automotive and the Internet of Things (IoT), TSMC is implementing artificial intelligence (AI) and machine learning technologies, and has developed systems for Precise Fault Detection and Classification System, Intelligent Advanced Equipment Control and Intelligent Advanced Process Control to ensure the consistency of tool matching and process stability. Combined Intelligent Process Variation Detection with foundry know-how to identify the potential defect and minimize the convergence of process variation through self-diagnosis and cognitive action. As the result, each chip can be precisely controlled at the nanometer level to produce the highest quality wafers for customers.