Abstract
In-line anomaly detection (AD) not only identifies the needs for semiconductor equipment maintenance but also indicates potential line yield problems. Prompt AD based on available equipment sensory data (ESD) facilitates proactive yield and operations management. However, ESD items are highly diversified and drastically scale up along with the increased use of sensors. Even veteran engineers lack knowledge about ESD items for automated AD. This paper presents a novel Spectral and Time Autoencoder Learning for Anomaly Detection (STALAD) framework. The design consists of four innovations: (1) identification of cycle series and spectral transformation (CSST) from ESD, (2) unsupervised learning from CSST of ESD by exploiting Stacked AutoEncoders, (3) hypothesis test for AD based on the difference between the learned normal data and the tested sample data, (4) dynamic procedure control enabling periodic and parallel learning and testing. Applications to ESD of an HDP-CVD tool demonstrate that STALAD learns normality without engineers’ prior knowledge, is tolerant to some abnormal data in training input, performs correct AD, and is efficient and adaptive for fab applications. Complementary to the current practice of using control wafer monitoring for AD, STALAD may facilitate early detection of equipment anomaly and assessment of impacts to process quality.
Highlights
In-line equipment anomaly detection (AD) identifies unusual behaviors in equipment sensory data (ESD) [1], which estimates the needs for equipment maintenance or repair and affects process control for end-of-line yield
Identification of cycle series hidden in ESD and spectral transformation of cycle series, Unsupervised learning of normal features from cycle series and their spectral transformation by exploiting Stacked AutoEncoders, One-tail hypothesis test for anomaly detection based on the difference between the learned normal sequence and the tested sample sequence, and Dynamic procedure control which coordinates both learning and testing phases to enable periodic and parallel executions of learning and testing
The apparent periodicity shown in a cycle series series (CS) inspires two innovative designs of the data preprocessing module: (1) identifications of CSes hidden in ESD, and (2) spectral transformation of sample sequences in a CS to characterize its periodic features
Summary
In-line equipment anomaly detection (AD) identifies unusual behaviors in equipment sensory data (ESD) [1], which estimates the needs for equipment maintenance or repair and affects process control for end-of-line yield. Sample sequences of various SVIDs differ in time series patterns or features, they are generated during tool processing of one same recipe. Equipment engineers can monitor sample sequences from various SVIDs of individual tools for anomaly detection. There has been research work on general equipment AD by using ESD in time series [11] and some on semiconductor manufacturing tools in specific [12]. Applied a moving window principal component analysis model with recipe information to capture the normal model of ESD and detect potential tool anomalies. To address the four problems and resolve their challenges, we shall propose a novel framework of spectral and time autoencoder learning for anomaly detection (STALAD).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
