The paper introduces the Dynamic Multi-Criteria Bottleneck Severity Index (DMBSI), a data-driven method for ranking bottleneck severity in reentrant wafer-fabrication systems. It combines multiple cycle-time signals, process-parameter changes, and rework effects into an interpretable score, with a time-windowed component for detecting bottleneck migration. On MES logs from 22 production lots at a commercial 200 mm Seagate Technology wafer line, GA-optimised DMBSI reached Pearson r = 0.80 with observed cycle-time contributions, versus 0.74 for an expert heuristic, 0.60 for Theory of Constraints, and -0.30 for Value Stream Mapping.
No heat snapshots are available in the last 24 hours.