NeFut Logo NeFut
Admin Login

[CS.AI] Dynamic Multi-Criteria Bottleneck Severity Index for Semiconductor Wafer Manufacturing

Published at: 2026-07-30 22:00 Last updated: 2026-07-30 23:39
#algorithm #optimization #C++

Wafer fabrication exhibits unique characteristics, including reentrant process flows, variable bottlenecks, and highly variable process conditions. To identify the most severe bottleneck at each moment in semiconductor wafer manufacturing, this research presents the Dynamic Multi-Criteria Bottleneck Severity Index (DMBSI), a new data-driven methodology that analyzes multiple diagnostic signals of cycle time, the impact of changes in process parameters, and the effect of reworks on cycle time to generate an interpretable, unified measure of bottleneck severity.

The experimental validation of DMBSI was conducted using Manufacturing Execution System (MES) logs collected from 22 wafer production lots at a commercial 200 mm wafer fabrication line operated by Seagate Technology. Using 5-fold cross-validation, the GA-optimized DMBSI achieved a Pearson correlation of r = 0.80 with observed cycle-time contributions, representing an 8.1% improvement over the expert heuristic baseline (r = 0.74) and substantially outperforming the Theory of Constraints (TOC; r = 0.60) and Value Stream Mapping (VSM; r = -0.30).

Furthermore, the unique time-windowed component of DMBSI enabled the identification of temporal bottleneck migration patterns, shifting from dominant constraints associated with dielectric deposition steps in early production windows to those related to over- and under-inspection in later production windows. The integrated what-if counterfactual analysis demonstrated that a 50% reduction in waiting time at the top-ranked bottleneck step would reduce the mean cycle time by 7.2%, with the top five bottleneck steps offering a combined potential reduction of approximately 19%.

Blogger's Review: This paper provides an innovative bottleneck analysis tool for semiconductor manufacturing through DMBSI, significantly enhancing production efficiency and cycle time predictability with a genetic algorithm optimization approach. Its practicality and accuracy offer valuable insights for process optimization in other industries, making it worthy of attention.

Original Source: https://arxiv.org/abs/2607.24819

[h] Back to Home