A Data-Driven Approach for Real-Time Bottleneck Detection and Optimization in Semiconductor Manufacturing Using Active Period Method and Visualization

With the rapid development of the semiconductor industry, identifying and optimizing bottlenecks is crucial for improving production line efficiency. This paper proposes a method combining Activity Cycle Method (APM) and data visualization techniques. APM identifies key bottlenecks in semiconductor manufacturing by analyzing the continuous uptime of machines and the duration of their activity cycles. Data visualization tools are then used to present these key bottlenecks in an intuitive and actionable manner. Applying both methods to a real-world semiconductor manufacturing environment significantly improves production efficiency and machine utilization, making this method practically applicable in semiconductor manufacturing.