data collection

Data collection automatically gathers process data and metrology results via automation systems, enabling SPC, traceability, and advanced analytics. Data types: (1) Summary data—single values per wafer/lot (average CD, film thickness, particle count); (2) Trace data—time-series sensor data during processing (high-frequency, high-volume); (3) Event data—discrete occurrences (wafer start, process complete, alarms); (4) Context data—lot ID, recipe, tool chamber, slot. SECS/GEM data collection: Stream 6 (S6F11 event report, S6F15 event report with data). EDA/Interface A: modern high-speed data interface for trace data (E164 standard). Data collection setup: define collection events (triggers), define report contents (which parameters), define trace triggers and parameters. Data volume considerations: trace data can generate GB/day—selective collection and compression essential. Data flow: Equipment → EDA module → Historian/Data warehouse → Analytics applications. Applications: (1) SPC—monitor key parameters; (2) FDC—fault detection from trace signatures; (3) Traceability—relate wafer history to final yield; (4) Process engineering—troubleshooting and optimization; (5) Virtual metrology—predict measurements from sensor data. Data quality: timestamp accuracy, sensor calibration, complete collection (no gaps). Foundation for data-driven manufacturing, yield improvement, and Industry 4.0 smart fab initiatives.