Cross-Process Defect Attribution using Potential Loss Analysis

TL;DR

This paper introduces Potential Loss Analysis (PLA), a novel framework for cross-process wafer defect root cause analysis that improves attribution accuracy by solving a Bellman equation for optimal outcomes.

Contribution

The paper presents PLA, a new method that enhances defect attribution by integrating outcome prediction and root cause analysis through a unified Bellman equation approach.

Findings

PLA effectively attributes wafer defects to upstream processes.

The framework accurately predicts defect densities using real wafer data.

PLA outperforms previous methods in defect attribution accuracy.

Abstract

Cross-process root-cause analysis of wafer defects is among the most critical yet challenging tasks in semiconductor manufacturing due to the heterogeneity and combinatorial nature of processes along the processing route. This paper presents a new framework for wafer defect root cause analysis, called Potential Loss Analysis (PLA), as a significant enhancement of the previously proposed partial trajectory regression approach. The PLA framework attributes observed high wafer defect densities to upstream processes by comparing the best possible outcomes generated by partial processing trajectories. We show that the task of identifying the best possible outcome can be reduced to solving a Bellman equation. Remarkably, the proposed framework can simultaneously solve the prediction problem for defect density as well as the attribution problem for defect scores. We demonstrate the effectiveness of the proposed framework using real wafer history data.