Sequence-Aware Inline Measurement Attribution for Good-Bad Wafer Diagnosis

TL;DR

This paper introduces Trajectory Shapley Attribution (TSA), a novel method extending Shapley values to account for process sequence, improving root cause analysis in complex semiconductor wafer defect diagnosis.

Contribution

The paper presents TSA, a new attribution framework that incorporates process sequence information, addressing limitations of traditional Shapley values in manufacturing diagnostics.

Findings

TSA effectively identifies key measurement items linked to wafer defects.

TSA outperforms standard Shapley values in sequential process analysis.

Application to real-world data demonstrates improved diagnosis accuracy.

Abstract

How can we identify problematic upstream processes when a certain type of wafer defect starts appearing at a quality checkpoint? Given the complexity of modern semiconductor manufacturing, which involves thousands of process steps, cross-process root cause analysis for wafer defects has been considered highly challenging. This paper proposes a novel framework called Trajectory Shapley Attribution (TSA), an extension of Shapley values (SV), a widely used attribution algorithm in explainable artificial intelligence research. TSA overcomes key limitations of standard SV, including its disregard for the sequential nature of manufacturing processes and its reliance on an arbitrarily chosen reference point. We applied TSA to a good-bad wafer diagnosis task in experimental front-end-of-line processes at the NY CREATES Albany NanoTech fab, aiming to identify measurement items (serving as proxies for process parameters) most relevant to abnormal defect occurrence.