Semiconductor Manufacturing Lot Quality Prediction via Frilled Lizard Optimization and Graph Attention Fusion Network

Paper ID: EIJTEM_2018_5_3_28-35

Author's Name: Dileep Kumar Ghattamaneni

Volume: 5

Issue: 3

Year: 2018

Page No: 28-35

Abstract:

In semiconductor manufacturing, accurate and timely lot quality prediction is critical for minimizing scrap rates and optimizing production efficiency. Traditional quality assessment methods often rely on expensive, time-intensive equipment and limited product sampling, which can result in undetected quality issues and yield losses. While existing approaches such as RGRN, FDC-CNN, and GMDH-PNN have made progress in modeling process variables, they typically lack the ability to effectively capture temporal alarm dynamics and often do not integrate equipment health data comprehensively. To address these limitations, this work proposes a novel Probabilistic Spiking Neural Networks with Tactical Unit Algorithm (PSNN-TUA) framework for robust lot quality prediction. Alarm logs and scrap reports collected over a year are pre-processed and transformed into structured matrices representing frequency, action, and severity, which are fused using a custom weighting scheme to emphasize high-risk events. Alarms are categorized based on quality impact, and the Analytical Clifford Fourier Mellin Transform (ACFMT) is used for robust feature extraction, ensuring invariance to geometric transformations. The PSNN captures temporal alarm patterns, while TUA adaptively optimizes hyperparameters through a bio-inspired search strategy. The proposed model outperforms existing methods with a high accuracy of 97.84%, demonstrating superior reliability and precision in real-time semiconductor quality monitoring.

Keywords: Analytical Clifford Fourier Mellin Transform, Lot quality prediction, Probabilistic Spiking Neural Networks, Semiconductor manufacturing, Tactical Unit Algorithm

References:

1. Lu, H.P. and Weng, C.I., 2018. Smart manufacturing technology, market maturity analysis and technology roadmap
in the computer and electronic product manufacturing industry. Technological Forecasting and Social Change, 133,
pp.85-94.
2. Farooq, M.A., Kirchain, R., Novoa, H. and Araujo, A., 2017. Cost of quality: Evaluating cost-quality trade-offs for
inspection strategies of manufacturing processes. International Journal of Production Economics, 188, pp.156-
166..
3. Villarini, M., Cesarotti, V., Alfonsi, L. and Introna, V., 2017. Optimization of photovoltaic maintenance plan by
means of a FMEA approach based on real data. Energy Conversion and Management, 152, pp.1-12.
4. Wang, K., Gopaluni, R.B., Chen, J. and Song, Z., 2018. Deep learning of complex batch process data and its
application on quality prediction. IEEE Transactions on Industrial Informatics, 16(12), pp.7233-7242.
5. Syafrudin, M., Fitriyani, N.L., Li, D., Alfian, G., Rhee, J. and Kang, Y.S., 2017. An open source-based real-time
data processing architecture framework for manufacturing sustainability. Sustainability, 9(11), p.2139.
6. Tello, G., Al-Jarrah, O.Y., Yoo, P.D., Al-Hammadi, Y., Muhaidat, S. and Lee, U., 2018. Deep-structured machine
learning model for the recognition of mixed-defect patterns in semiconductor fabrication processes. IEEE
Transactions on Semiconductor Manufacturing, 31(2), pp.315-322.
7. Lee, K.B., Cheon, S. and Kim, C.O., 2017. A convolutional neural network for fault classification and diagnosis in
semiconductor manufacturing processes. IEEE Transactions on Semiconductor Manufacturing, 30(2), pp.135-142.
8. Jia, X., Di, Y., Feng, J., Yang, Q., Dai, H. and Lee, J., 2018. Adaptive virtual metrology for semiconductor
chemical mechanical planarization process using GMDH-type polynomial neural networks. Journal of Process
Control, 62, pp.44-54.
9. Affes, M., Meziou, M.S. and Ghorbel, F., 2017, December. A new watermarking method based on analytical clifford
Fourier Mellin transform. In International Workshop on Representations, Analysis and Recognition of Shape and
Motion From Imaging Data (pp. 181-191). Cham: Springer International Publishing.
10. Lee, J.H., Delbruck, T. and Pfeiffer, M., 2016. Training deep spiking neural networks using
backpropagation. Frontiers in neuroscience, 10, p.508.
11. Dellaert, N. and Jeunet, J., 2017. A variable neighborhood search algorithm for the surgery tactical planning
problem. Computers & Operations Research, 84, pp.216-225.

View PDF