Anomaly detection and remedial recommendation
AssigneeInternational Business Machines Corporation
InventorsDzung Phan, Robert Baseman, Nam H. Nguyen, Fateh Tipu, Ramachandran Muralidhar
Life of this patent
Estimating the term…
Abstract
Anomaly detection and remedial recommendation techniques for improving the quality and yield of microelectronic products are provided. In one aspect, a method for quality and yield improvement via anomaly detection includes: collecting time series sensor data during individual steps of a semiconductor manufacturing process; calculating anomaly scores for each of the individual steps using a predictive model; and implementing changes to the semiconductor manufacturing process based on the anomaly scores. A system for quality and yield improvement via anomaly detection is also provided.
Claims15
A method for quality and yield improvement via anomaly detection, the method comprising:
- collecting time series sensor data during individual steps of a semiconductor manufacturing process being performed using a semiconductor fabrication tool;
- calculating anomaly scores for each of the individual steps using a predictive model;
- and
- implementing changes to the semiconductor manufacturing process by automatically adjusting the semiconductor fabrication tool for one or more of the individual steps that, based on the anomaly scores, are anomalous,
- wherein the semiconductor fabrication tool has multiple processing chambers for performing each of the individual steps of the semiconductor manufacturing process, wherein the anomaly scores for at least one of the individual steps performed in at least one of the processing chambers exceed a pre-determined threshold value, and wherein the method further comprises:
- recommending remedial actions for the individual steps having the anomaly scores that exceed the pre-determined threshold value;
- and
- implementing the remedial actions in the semiconductor fabrication tool.
A computer program product for quality and yield improvement via anomaly detection, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to:
- collect time series sensor data during individual steps of a semiconductor manufacturing process being performed using a semiconductor fabrication tool;
- calculate anomaly scores for each of the individual steps using a predictive model, wherein the predictive model comprises a two-layer sparse Gaussian mixture model;
- and
- implement changes to the semiconductor manufacturing process by automatically adjusting the semiconductor fabrication tool for one or more of the individual steps that, based on the anomaly scores, are anomalous,
- wherein the semiconductor fabrication tool has multiple processing chambers for performing each of the individual steps of the semiconductor manufacturing process, wherein the anomaly scores for at least one of the individual steps performed in at least one of the processing chambers exceed a pre-determined threshold value, and wherein the program instructions further cause the computer to:
- recommend remedial actions for the individual steps having the anomaly scores that exceed the pre-determined threshold value;
- and
- implement the remedial actions in the semiconductor fabrication tool by automatically adjusting at least one parameter of the processing chambers in which the individual steps having the anomaly scores that exceed the pre-determined threshold value are performed based on the remedial actions recommended.
A system for quality and yield improvement via anomaly detection, comprising:
- a sensor network for collecting time series sensor data during individual steps of a semiconductor manufacturing process being performed using a semiconductor fabrication tool;
- an anomaly score calculator for calculating anomaly scores for each of the individual steps using a predictive model, wherein the predictive model comprises a two-layer sparse Gaussian mixture model;
- and
- a process operations update module for implementing changes to the semiconductor manufacturing process by automatically adjusting the semiconductor fabrication tool for one or more of the individual steps that, based on the anomaly scores, are anomalous,
- wherein the semiconductor fabrication tool has multiple processing chambers for performing each of the individual steps of the semiconductor manufacturing process, wherein the anomaly scores for at least one of the individual steps performed in at least one of the processing chambers exceed a threshold value, and wherein the process operations update module recommends remedial actions for the individual steps having the anomaly scores that exceed the pre-determined threshold value which are implemented in the semiconductor fabrication tool by automatically adjusting at least one parameter of the processing chambers in which the individual steps having the anomaly scores that exceed the pre-determined threshold value are performed based on the remedial actions recommended.
Dependencies are inferred from the claim text.