Feature Learning menggunakan Deep Variational Autoencoder untuk prediksi cacat pada mesin mobil

• NANANG SUSANTO
• 14210248

---

ABSTRAK

ABSTRAK

Nama              : Nanang Susanto
NIM                 : 14210248
Program Studi : Ilmu Komputer
Fakultas          : Teknologi Informasi
Jenjang           : Strata Dua (S2)
Konsentrasi     : Data Mining
Judul      : "Feature Learning menggunakan Deep Variational Autoencoder untuk prediksi cacat pada mesin mobil"

Industri otomotif menghadapi tantangan dalam memastikan kualitas mesin mobil yang diproduksi. Kelemahan indera manusia untuk mendeteksi cacat pada mesin memiliki keterbatasan. Permasalahan krusial yang muncul adalah deteksi cacat pada mesin. Data cacat yang terkumpul seringkali berjumlah besar membuat deteksi secara manual menjadi tidak efisien dan tidak akurat. Penelitian ini menggunakan dataset kualitas engine sebanyak 224.239 data dengan 90 fitur. Tahap pre-processing data meliputi perbaikan data missing value, lalu dilakukan feature correlation dengan metode pearson, dan memilih fitur yang digunakan. Setelah itu, dilakukan label encoder, dan standard scaler. Pada proses eksperimen diawali membuat baseline, lalu dilanjutkan dengan perbaikan data imbalance menggunakan SMOTE, dan rekonstruksi fitur menggunakan Autoencoder dan Variational Autoencoder. Penelitian ini menggunakan algoritma CNN, MLP dan LSTM. Hasil penelitian menunjukkan metode SMOTE dapat meningkatkan performa keseluruhan model. Hasil terbaik pada penelitian ini adalah menggunakan algoritma CNN-SMOTE-Variational Autoencoder, dimana mendapat precission sebesar 99,63%. Hal tersebut menunjukkan, metode rekonstruksi dimensi Variational Autoencoder bisa mengatasi prediksi cacat pada data cacat mesin mobil dengan kondisi imbalance data. Dengan hal tersebut diharapkan proses pengecekan cacat pada industri otomotif menjadi lebih baik dan efisien.

KATA KUNCI

Deep Learning,CNN,Variational Autoencoder,SMOTE,Cacat Produk

---

DAFTAR PUSTAKA

?DAFTAR REFERENSI

[1] J. hye Jun, T. W. Chang, and S. Jun, "Quality prediction and yield improvement in process manufacturing based on data analytics," Processes, vol. 8, no. 9, Sep. 2020, doi:10.3390/pr8091068.

[2] A. Insan Waluya, M. Ali Iqbal, and R. Indradewa, "How product quality, brand image, and customer satisfaction affect the purchase decisions of Indonesian automotive customers," 2019. [Online]. Available: http://www.marklines.com

[3] "data gaikindo," Penjualan Mob. Nas. Akhir Tahun 2022, vol. 1, no. 1, p. 4, 2022, [Online]. Available: https://www.gaikindo.or.id/gaikindo-penjualan-
mobil-nasional-tembus-950-ribu-unit-di-akhir-tahun-2022/

[4] Detik, "recall daihatsu," Recall Daihatsu Rocky, 2022.
https://oto.detik.com/mobil/catat-recall-daihatsu-rocky/

[5] "recall quality," Sekilas Mengen. Recall Prod. Kendaraan dan Regulasinya di Indones. -GAIKINDO, vol. 1, no. 1, p. 2, 2021.

[6] "2019201QS20Release_Final".

[7] P. Tubaro and A. A. Casilli, "Micro-work, artificial intelligence and the automotive industry," J. Ind. Bus. Econ., vol. 46, no. 3, pp. 333-345, Sep. 2019, doi: 10.1007/s40812-019-00121-1.

[8] L. Li, K. Ota, and M. Dong, "Deep Learning for Smart Industry: Efficient Manufacture Inspection System with Fog Computing," IEEE Trans. Ind. Informatics, vol. 14, no. 10, pp. 4665-4673, 2018, doi: 10.1109/TII.2018.2842821.

[9] J. Yang, S. Li, Z. Wang, H. Dong, J. Wang, and S. Tang, "Using deep learning to detect defects in manufacturing: A comprehensive survey and current challenges," Materials, vol. 13, no. 24. MDPI AG, pp. 1-23, Dec. 02, 2020. doi: 10.3390/mal3245755.

[10] N. Azeem and S. Usmani, "Defect Prediction Leads to High Quality 

Product," J. Softw. Eng. Appl., vol. 04, no. 11, pp. 639-645, 2011, doi: 10.4236/jsea.2011.411075.

[II] H. Pan, Z. Pang, Y. Wang, Y. Wang, and L. Chen, "A New Image Recognition and Classification Method Combining Transfer Learning Algorithm and MobileNet Model for Welding Defects," IEEE Access, vol. 8, pp. 119951-119960, 2020, doi: 10.1109/ACCESS.2020.3005450.

[12] S. Luo, X. Ma, J. Xu, M. Li, and L. Cao, "Deep leaming based monitoring of spatter behavior by the acoustic signal in selective laser melting," Sensors, vol. 21, no. 21, Nov. 2021, doi: 10.3390/s21217179.

[13] J. Soviar, M. Holub?ík, J. Vodák, M. Rechtorik, and F. Pollák, "The presentation of automotive brands in the on-line environment-the perspective of KIA, Peugeot, Toyota and VW in the Slovak Republic," Sustain., vol. 11, no. 7, 2019, doi: 10.3390/SUI 1072132.

[14] F. S. Global, "A new life-cycle assessment of the greenhouse gas emissions of combustion engine and electric passenger cars in major markets," Communications, no. July, 2021, [Online]. Available: https://www.team- bhp.com/forum/attachments/electric-cars/2310712d1653112667-things- consider-when-buying-electric-car-global-lea-passenger-cars-jul2021-0.pdf

[15] I. Paul, "Dimensionality Reduction for Microarray Data: An Analytical Survey." [Online]. Available: www.ijdieret.in

[16] P. Park, P. Di Marco, H. Shin, and J. Bang, "Fault detection and diagnosis using combined autoencoder and long short-term memory network," Sensors (Switzerland), vol. 19, no. 21, pp. 1-17, 2019, doi: 10.3390/s19214612.

[17] S. Liu, H. Jiang, Z. Wu, and X. Li, “Rolling bearing fault diagnosis using variational autoencoding generative adversarial networks with deep regret analysis," Meas. J. Int. Meas. Confed., vol. 168, no. July 2020, p. 108371, 2021, doi: 10.1016/j.measurement. 2020.108371.

[18] D. Detection, "Defect detection," Engineering, vol. 251, no. 2, p. 24, 2010, doi: 10.1201/b15870-17.

[19] F. Ulger, S. E. Yuksel, and A. Yilmaz, "Anomaly Detection for Solder
Joints Using $\beta$-VAE," Apr. 2021, doi: 10.1109/TCPMT 2021.3121265.

[20] Q. Li and Z. Song, "Ensemble-Learning-Based Prediction of Steel Bridge Deck Defect Condition," Appl. Sci., vol. 12, no. 11, 2022, doi: 10.3390/app12115442.

[21] C. Knaak, J. von Eßen, M. Kröger, F. Schulze, P. Abels, and A. Gillner, "A spatio-temporal ensemble deep learning architecture for real-time defect detection during laser welding on low power embedded computing boards," Sensors, vol. 21, no. 12, Jun. 2021, doi: 10.3390/s21124205.

[22] A. H. Ombabi, W. Ouarda, and A. M. Alimi, "Deep learning CNN-LSTM framework for Arabic sentiment analysis using textual information shared in social networks," Soc. Netw. Anal. Min., vol. 10, no. 1, Dec. 2020, doi: 10.1007/s13278-020-00668-1.

[23] W. S. Albaldawi and R. M. Almuttairi, "Hybrid ANOVA and LASSO Methods for Feature Selection and Linear Support Vector, Multilayer Perceptron and Random Forest Classifiers Based on Spark Environment for Microarray Data Classification," IOP Conf. Ser. Mater. Sci. Eng., vol.1094, no. 1, p. 012107, Feb. 2021, doi: 10.1088/1757-899x/1094/1/012107.

[24] J. Lee, Y. C. Lee, and J. T. Kim, "Fault detection based on one-class deep learning for manufacturing applications limited to an imbalanced
database," J. Manuf. Syst., vol. 57, pp. 357-366, Oct. 2020, doi: 10.1016/j.jmsy.2020.10.013.

[25] S. K. Mohapatra, J. K. Swain, and M. N. Mohanty, Detection of Diabetes Using Multilayer Perceptron, vol. 846. Springer Singapore, 2019. doi: 10.1007/978-981-13-2182-5_11.

[26] T. Ji and N. Mohamad Nor, "Deep Learning-Empowered Digital Twin Using Acoustic Signal for Welding Quality Inspection," Sensors, vol. 23, no. 5, Mar. 2023, doi: 10.3390/s23052643.

[27] T. B. Kimber, Y. Chen, and A. Volkamer, "Deep learning in virtual screening: Recent applications and developments," Int. J. Mol. Sci., vol. 22, no. 9, 2021, doi: 10.3390/ijms22094435.

[28] W. Deelder et al., "Machine learning predicts accurately mycobacterium tuberculosis drug resistance from whole genome sequencing data," Front. Genet., vol. 10, no. SEP, Sep. 2019, doi: 10.3389/fgene.2019.00922.

[29] G. Li, Q. Liu, S. Zhao, W. Qiao, and X. Ren, "Automatic crack recognition for concrete bridges using a fully convolutional neural network and naive Bayes data fusion based on a visual detection system," Meas. Sci. Technol., vol. 31, no. 7, Jul. 2020, doi: 10.1088/1361-6501/ab79c8.

[30] N. U. Din, L. Zhang, and Y. Yang, "Automated Battery Making Fault Classification Using Over-Sampled Image Data CNN Features," Sensors, vol. 23, no. 4, Feb. 2023, doi: 10.3390/s23041927.

[31] X. Lv, F. Duan, J. J. Jiang, X. Fu, and L. Gan, "Deep metallic surface defect detection: The new benchmark and detection network," Sensors (Switzerland), vol. 20, no. 6, Mar. 2020, doi: 10.3390/20061562.

[32] O. Hrizi et al., "Tuberculosis Disease Diagnosis Based on an Optimized Machine Learning Model," J. Healthc. Eng., vol. 2022, 2022, doi: 10.1155/2022/8950243.

[33] J. Brownlee, "What is the Difference Between a Batch and an Epoch in a Neural Network?," Mach. Learn. Mastery, no. July, pp. 3-4, 2018.

[34] B. Adelmann, M. Schleier, and R. Hellmann, "Laser cut interruption detection from small images by using convolutional neural network," Sensors (Switzerland), vol. 21, no. 2, pp. 1-13, Jan. 2021, doi: 10.3390/s21020655.

[35] J. C. Parducci, "Deep Learning Object-Based Detection of Manufacturing Defects Deep Learning Object-Based Detection of Manufacturing Defects in X-ray Inspection Imaging in X-ray Inspection Imaging", doi: 

10.25777/td75-fw04.

[36] K. Nirmalakumari, H. Rajaguru, and P. Rajkumar, "Performance analysis of classifiers for colon cancer detection from dimensionality reduced microarray gene data," Int. J. Imaging Syst. Technol., vol. 30, no. 4, pp. 1012-1032, Dec. 2020, doi: 10.1002/ima.22431.

[37] W. Zhang, L. Niu, D. Zhang, G. Wang, F. U. D. Farrukh, and C. Zhang, "HW-ADAM: FPGA-Based Accelerator for Adaptive Moment Estimation," Electron., vol. 12, no. 2, 2023, doi: 10.3390/electronics 12020263.

[38] A. Alacaoglu, Y. Malitsky, P. Mertikopoulos, and V. Cevher, "A new regret analysis for Adam-Type algorithms," 37th Int. Conf. Mach. Learn. ICML 2020, vol. PartF16814, no. 1, pp. 179–187, 2020.

[39] D. Zou, Y. Cao, D. Zhou, and Q. Gu, "Gradient descent optimizes over- parameterized deep ReLU networks," Mach. Learn., vol. 109, no. 3, pp. 467-492, 2020, doi: 10.1007/s10994-019-05839-6.

[40] S. Cheon, H. Lee, C. O. Kim, and S. H. Lee, "Convolutional Neural Network for Wafer Surface Defect Classification and the Detection of Unknown Defect Class," IEEE Trans. Semicond. Manuf., vol. 32, no. 2, pp. 163-170, May 2019, doi: 10.1109/TSM.2019.2902657.

[41] D. M. Tsai and P. H. Jen, "Autoencoder-based anomaly detection for surface defect inspection," Adv. Eng. Informatics, vol. 48, no. February, p. 101272, 2021, doi: 10.1016/j.aci 2021.101272.

[42] J. K. Chow, Z. Su, J. Wu, P. S. Tan, X. Mao, and Y. H. Wang, "Anomaly detection of defects on concrete structures with the convolutional autoencoder," Adv. Eng. Informatics, vol. 45, no. December 2019, 2020, doi: 10.1016/j.aci.2020.101105.

[43] A. Ashfahani, M. Pratama, E. Lughofer, and Y. S. Ong, "DEVDAN: Deepevolving denoising autoencoder," Neurocomputing, vol. 390, no. xxxx, pp. 297-314, 2020, doi: 10.1016/j.neucom 2019.07.106.

[44] A. He and X. Jin, "Deep Variational Autoencoder Classifier for Intelligent Fault Diagnosis Adaptive to Unseen Fault Categories," IEEE Trans. Reliab., vol. 70, no. 4, pp. 1581-1595, 2021, doi: 10.1109/TR.2021.3090310.

[45] R. Guo, H. Liu, G. Xie, and Y. Zhang, "Weld Defect Detection from Imbalanced Radiographic Images Based on Contrast Enhancement Conditional Generative Adversarial Network and Transfer Learning," IEEE Sens. J., vol. 21, no. 9, pp. 10844-10853, May 2021, doi: 10.1109/JSEN.2021.3059860.

[46] L. Chen, J. Hua, and X. He, "Coexpression Network Analysis-Based Identification of Critical Genes Differentiating between Latent and Active Tuberculosis," Dis. Markers, vol. 2022, 2022, doi: 10.1155/2022/2090560.

[47] Y. Luo et al., "Development of diagnostic algorithm using machine learning for distinguishing between active tuberculosis and latent tuberculosis infection," BMC Infect. Dis., vol. 22, no. 1, Dec. 2022, doi: 10.1186/s12879-022-07954-7.

[48] I. F. Ruslan, M. F. Alby, and M. Lubis, "Applying Data Governance using DAMA-DMBOK 2 Framework: The Case for Human Capital Management Operations," ACM Int. Conf. Proceeding Ser., pp. 336-342, 2022, doi: 10.1145/3568834.3568866.

[49] O. S. Amosov, S. G. Amosova, and I. O. lochkov, "Deep Neural Network Recognition of Rivet Joint Defects in Aircraft Products," Sensors, vol. 22, no. 9, May 2022, doi: 10.3390/s22093417.

[50] J. Wang, X. Zhou, and J. Wu, "Chip appearance defect recognition based on convolutional neural network," Sensors, vol. 21, no. 21, Nov. 2021, doi: 10.3390/s21217076.

[51] A. Bhattacharya and S. G. Cloutier, "End-to-end deep learning framework for printed circuit board manufacturing defect classification," Sci. Rep., vol. 12, no. 1, Dec. 2022, doi: 10.1038/s41598-022-16302-3.

[52] W. Dai et al., "Deep learning assisted vision inspection of resistance spot welds," J. Manuf. Process., vol. 62, pp. 262-274, Feb. 2021, doi: 10.1016/j.jmapro.2020.12.015.

[53] J. Zeng, B. Cao, and R. Tian, "Quality monitoring for micro resistance spot welding with class-imbalanced data based on anomaly detection," Appl. Sci., vol. 10, no. 12, p. 4204, 2020, doi: 10.3390/APP10124204.

[54] Joon-Hyung Park, "Analysis of Training Deep Learning Models for PCB Defect Detection," Engineering, vol. 251, no. 2, p. 24, 2023, doi: 10.1201/b15870-17.

[55] L. Posilovi?, D. Medak, F. Milkovi?, M. Subaši?, M. Budimir, and S. Lon?ari?, "Deep learning-based anomaly detection from ultrasonic images," Ultrasonics, vol. 124, p. 106737, Aug. 2022, doi: 10.1016/J.ULTRAS.2022.106737.