Integrated Deep Learning and Statistical Process Control for Online Monitoring of Manufacturing Processes

Ahmad, Safwan; Enshaei, Nastaran; Naderkhani, Farnoosh; Awasthi, Anjali

doi:10.1109/icphm49022.2020.9187046

Cited by 2 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…6). They are notably used for object tracking in autonomous vehicles or cell classification for cancer detection, but also in semiconductor industrial process control to improve accuracy in default detection [5,6]. They are also able to adapt themselves better to contrast/brightness and product/design variation than the mathematical image treatments commonly used by metrology and defectivity equipments.…”

Section: Region-based Convolutional Neural Networkmentioning

confidence: 99%

“…To train those neural networks, a dataset consisting of images containing the objects to be detected, and an annotation file containing the location, size, and class of said objects must be provided. In our case, to reduce the amount of data and computing time needed for this training, transfer learning from a pre-trained network (RetinaNet [5]) was used. RetinaNet is one of the best one-stage object detection models that has proven to work well with dense and small-scale objects.…”

Section: Region-based Convolutional Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Full Wafer Process Control Through Object Detection Using Region-Based Convolutional Neural Networks

Alcaire

Cunff

Tortai

et al. 2022

2022 33rd Annual SEMI Advanced Semiconductor Manufacturing Conference (ASMC)

View full text Add to dashboard Cite

Full wafer measurement techniques are used in the semiconductor industry to acquire information at a large scale to control process variation or detect potential defects. This process usually results in the generation of full wafer images, containing various objects that need to be identified to evaluate their impact on the final product performance. Artificial intelligence is very powerful to automate this identification routine. In this paper, we present the application of Region-based Convolutional Neural Networks (RCNN) for enhanced process control from full wafer images gathered by two industrial metrology equipments.

show abstract

Section: Region-based Convolutional Neural Networkmentioning

confidence: 99%

Section: Region-based Convolutional Neural Networkmentioning

confidence: 99%

Full Wafer Process Control Through Object Detection Using Region-Based Convolutional Neural Networks

Alcaire

Cunff

Tortai

et al. 2022

2022 33rd Annual SEMI Advanced Semiconductor Manufacturing Conference (ASMC)

View full text Add to dashboard Cite

show abstract

SSMSPC: self-supervised multivariate statistical in-process control in discrete manufacturing processes

et al. 2023

View full text Add to dashboard Cite

Self-supervised learning has demonstrated state-of-the-art performance on various anomaly detection tasks. Learning effective representations by solving a supervised pretext task with pseudo-labels generated from unlabeled data provides a promising concept for industrial downstream tasks such as process monitoring. In this paper, we present SSMSPC a novel approach for multivariate statistical in-process control (MSPC) based on self-supervised learning. Our motivation for SSMSPC is to leverage the potential of unsupervised representation learning by incorporating self-supervised learning into the general statistical process control (SPC) framework to develop a holistic approach for the detection and localization of anomalous process behavior in discrete manufacturing processes. We propose a pretext task called Location + Transformation prediction, where the objective is to classify both, the type and the location of a randomly applied augmentation on a given time series input. In the downstream task, we follow the one-class classification setting and apply the Hotelling’s $$T^2$$ T 2 statistic on the learned representations. We further propose an extension to the control chart view that combines metadata with the learned representations to visualize the anomalous time steps in the process data which supports a machine operator in the root cause analysis. We evaluate the effectiveness of SSMSPC with two real-world CNC-milling datasets and show that it outperforms state-of-the-art anomaly detection approaches, achieving $$100\%$$ 100 % and $$99.6\%$$ 99.6 % AUROC, respectively. Lastly, we deploy SSMSPC at a CNC-milling machine to demonstrate its practical applicability when used as a process monitoring tool in a running process.

show abstract

Integrated Deep Learning and Statistical Process Control for Online Monitoring of Manufacturing Processes

Cited by 2 publications

References 14 publications

Full Wafer Process Control Through Object Detection Using Region-Based Convolutional Neural Networks

Full Wafer Process Control Through Object Detection Using Region-Based Convolutional Neural Networks

SSMSPC: self-supervised multivariate statistical in-process control in discrete manufacturing processes

Contact Info

Product

Resources

About