Continuous Wavelet Transform and Siamese Network-Based Anomaly Detection in Multi-variate Semiconductor Process Time Series

TL;DR

This paper introduces a novel machine learning approach combining Continuous Wavelet Transform, fine-tuned VGG-16, and Siamese networks for effective anomaly detection in complex multi-variate semiconductor manufacturing time series data.

Contribution

It proposes a generic, image-based anomaly detection method using CWT and Siamese networks, addressing challenges like high dimensionality and class imbalance in semiconductor process data.

Findings

High accuracy in anomaly detection on real FAB data

Flexible application in supervised and semi-supervised settings

Effective comparison of signals using Siamese network embeddings

Abstract

Semiconductor manufacturing is an extremely complex process, characterized by thousands of interdependent parameters collected across diverse tools and process steps. Multi-variate time-series (MTS) analysis has emerged as a critical methodology for enabling real-time monitoring, fault detection, and predictive maintenance in such environments. However, anomaly prediction in semiconductor fabrication presents several critical challenges, including high data dimensionality, severe class imbalance due to the rarity of true faults, noisy and missing measurements, and non-stationary behavior of production systems. Furthermore, the complex interdependencies between variables and the delayed emergence of faults across downstream stages complicate both anomaly detection and root-cause-analysis. This paper presents a novel and generic approach for anomaly detection in MTS data using machine learning. The proposed methodology consists of three main steps: a) converting MTS data into image-based representations using the Continuous Wavelet Transform, b) developing a multi-class image classifier by fine-tuning a pretrained VGG-16 architecture on custom CWT image datasets, and c) constructing a Siamese network composed of two identical sub-networks, each utilizing the fine-tuned VGG-16 as a backbone. The network takes pairs of CWT images as input -one serving as a reference or anchor (representing a known-good signal), and the other as a query (representing an unknown signal). The model then compares the embeddings of both inputs to determine whether they belong to the same class at a given time step. Our approach demonstrates high accuracy in identifying anomalies on a real FAB process time-series dataset, offering a promising solution for offline anomaly detection in process and tool trace data. Moreover, the approach is flexible and can be applied in both supervised and semi-supervised settings.