PEMODELAN HYBRID SUPPORT VECTOR MACHINE DAN K-NEAREST NEIGBORS UNTUK PERAMALAN PERMINTAAN SUKU CADANG OTOMOTIF: PENDEKATAN MACHINE LEARNING TERHADAP DATA FLUKTUATIF
Article Sidebar
Published
Jan 15, 2026
Main Article Content
Abstract
Pengelolaan persediaan suku cadang otomotif menghadapi tantangan kompleks akibat pola permintaan yang fluktuatif, bersifat sporadis, dan tidak sepenuhnya mengikuti pola musiman. Penelitian ini bertujuan untuk mengevaluasi dan membandingkan kinerja empat pendekatan peramalan, yaitu Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Hybrid Weighted Ensemble (gabungan SVM dan KNN), serta Seasonal ARIMA (SARIMA). Dataset yang digunakan berasal dari riwayat penjualan suku cadang selama 33 bulan, dengan karakteristik permintaan yang sangat bervariasi antar item.Hasil evaluasi menunjukkan bahwa model KNN memberikan performa terbaik dengan nilai MAE 33,22, RMSE 35,92, MAPE 7,31%, dan sMAPE 7,63%, diikuti oleh SVM dengan MAE 62,36 dan sMAPE 14,84%. Model Hybrid menghasilkan performa menengah, sedangkan SARIMA menunjukkan akurasi terendah dengan sMAPE mencapai 28,23%. Visualisasi tren prediksi memperkuat temuan ini, di mana model berbasis machine learning mampu mengikuti fluktuasi aktual secara lebih konsisten dibandingkan pendekatan statistik klasik. Kebaruan penelitian ini terletak pada penerapan model hybrid yang secara khusus diadaptasikan untuk data permintaan suku cadang otomotif, serta integrasinya dengan teknik feature engineering seperti lag features dan seasonal encoding. Temuan ini merekomendasikan penggunaan pendekatan machine learning, khususnya KNN dan model hybrid, sebagai solusi praktis dalam sistem perencanaan persediaan. Implementasi model ini berpotensi menekan biaya penyimpanan, mengurangi risiko stockout maupun overstock, serta mendukung pengambilan keputusan berbasis data di tingkat operasional dan manajerial.
Article Details
How to Cite
[1]
B. M. Dinata and E. Ardhianto, “PEMODELAN HYBRID SUPPORT VECTOR MACHINE DAN K-NEAREST NEIGBORS UNTUK PERAMALAN PERMINTAAN SUKU CADANG OTOMOTIF: PENDEKATAN MACHINE LEARNING TERHADAP DATA FLUKTUATIF”, JURNAL EDUCATION AND DEVELOPMENT, vol. 14, no. 1, pp. 178-188, Jan. 2026.
Section
Artikel
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
References
Fatalifi, A. G. (2025). Pengembangan sistem peramalan permintaan menggunakan algoritma support vector regression untuk optimalisasi safety stock berbasis web (Studi kasus: JG). Nusaputra Repository. https://repository.nusaputra.ac.id/ 1441/1/AMERJID%20GHULAMSON%20FATALIFI%20%28Repository%29.pdf
Fatalifi, A. G. (2025). Pengembangan sistem peramalan permintaan menggunakan algoritma support vector regression untuk optimalisasi safety stock berbasis web. Nusaputra Repository. https://repository.nusaputra.ac.id/id/eprint/ 1441/1/AMERJID%20GHULAMSON%20FATALIFI%20%28Repository%29.pdf
Ardafan, F. H. B. (2024). Aplikasi invoice reader berbasis machine learning sebagai upaya optimasi proses bisnis.
Utama, J. P. (2022). The multi-criteria inventory model of aircraft spare part initial provisioning (A case study of Garuda Indonesia Airbus A330neo entry into service). Institut Teknologi Sepuluh Nopember.
Fatalifi, A. G. (2025). Pengembangan sistem peramalan permintaan menggunakan algoritma support vector regression untuk optimalisasi safety stock berbasis web (Studi kasus: JG). Nusaputra Repository. https://repository.nusaputra.ac.id/ 1441/1/AMERJID%20GHULAMSON%20FATALIFI%20%28Repository%29.pdf
Dewi, B. E. S., Haikal, S., & Sulistyowati, H. S. (2024). Penerapan machine learning menggunakan algoritma random forest untuk prediksi harga mobil bekas. Jurnal Teknologi dan Sistem Komputer, 12(2), 45–55. https://banisalehjurnal.ubs.ac.id/ index.php/tridi/article/download/38/42
Ramadhan, D. I. (2020). Analisis kinerja peramalan dan klasifikasi permintaan part otomotif dengan pendekatan time-series dan data mining. Institut Teknologi Sepuluh Nopember.
Prabowo, T., & Sari, N. (2023). Hybrid SVM and KNN approach for spare parts demand forecasting in the automotive industry. Jurnal Teknologi dan Sistem Komputer, 11(4), 201–210. https://jtsiskom.id/download/1104-prabowo2023.pdf
Nugroho, A. P., & Rahmawati, L. (2022). Integrating support vector machine and k-nearest neighbor algorithms to improve forecasting accuracy for automotive spare parts demand. Jurnal Ilmiah Teknik Industri, 21(2), 145–156. https://jiti.id/files/jiti2022-nugroho.pdf
Setiawan, D., & Lestari, M. (2021). Optimizing demand prediction of automotive spare parts using a hybrid machine learning model. Jurnal Sistem Informasi, 17(1), 33–44. https://jsi.id/download/jsi17-1-setiawan2021.pdf
Wulandari, E., & Pratama, R. (2024). Demand forecasting of motorcycle spare parts using machine learning algorithms. Jurnal Teknologi Informasi dan Ilmu Komputer, 11(2), 145–154. https://jtik.id/download/112-wulandari2024.pdf
Hidayat, M. R., & Fauzi, M. A. (2023). Predictive analytics for spare parts inventory management in the automotive sector using supervised learning. Jurnal Sistem Informasi, 19(3), 211–220. https://jsi.id/download/193-hidayat2023.pdf
Santoso, B., & Wijaya, A. (2022). Forecasting automotive spare parts demand with ensemble machine learning models. Jurnal Ilmu Komputer dan Informasi, 15(1), 57–66. https://jik.id/download/jik151-santoso2022.pdf
Yuliana, S., & Hakim, R. (2024). Optimization of spare parts inventory using machine learning-based demand forecasting. Jurnal Teknologi dan Sistem Komputer, 12(1), 88–97. https://jtsiskom.id/download/121-yuliana2024.pdf
Ramadhani, L. A., & Prasetyo, B. (2023). Hybrid forecasting model for spare parts stock optimization in automotive industry. Jurnal Sistem Informasi, 19(4), 299–308. https://jsi.id/download/194-ramadhani2023.pdf
Kurniawan, D., & Fitria, N. (2021). Application of support vector regression and k-nearest neighbor in spare parts inventory optimization. Jurnal Ilmiah Teknik Industri, 20(3), 177–186. https://jiti.id/files/jiti2021-kurniawan.pdf
Susanto, H., & Lestari, E. (2024). Automotive sales demand forecasting using hybrid SVM and KNN models. Jurnal Teknologi Informasi dan Ilmu Komputer, 12(2), 122–131. https://jtik.id/download/122-susanto2024.pdf
Maulana, R., & Saputra, A. (2023). Comparative analysis of SVM and KNN for demand prediction in the automotive industry. Jurnal Sistem Informasi, 19(2), 167–176. https://jsi.id/download/192-maulana2023.pdf
Anggraini, D., & Hidayat, T. (2021). Improving accuracy of vehicle demand forecasting using machine learning hybrid approach. Jurnal Ilmu Komputer dan Informasi, 14(4), 291–300. https://jik.id/download/jik144-anggraini2021.pdf
Putri, V. A., & Gunawan, F. (2024). Demand prediction for automotive supply chain using hybrid machine learning models. Jurnal Teknologi dan Sistem Komputer, 12(3), 201–210. https://jtsiskom.id/download/123-putri2024.pdf
Akbar, M. R., & Suryani, T. (2023). Machine learning-based forecasting for supply chain demand in the automotive sector. Jurnal Sistem Informasi, 19(1), 45–54. https://jsi.id/download/191-akbar2023.pdf
Dewantara, H., & Prasetia, A. (2021). Application of hybrid SVM-KNN models in automotive supply chain demand forecasting. Jurnal Ilmiah Teknik Industri, 20(1), 33–42. https://jiti.id/files/jiti2021-dewantara.pdf
Kusuma, V., & Budhidharma, V. (2025). Inventory management optimization in automotive spare parts industry using machine learning approach. Jurnal Ilmiah Multidisiplin, 5(1), 88–99. https://ejournal.seaninstitute.or.id/ index.php/esaprom/article/download/6537/5072
Sari, P. A., & Nugraha, R. (2024). Predictive analytics for automotive spare parts inventory using hybrid SVM and KNN. Jurnal Teknologi dan Sistem Komputer, 12(4), 277–286. https://jtsiskom.id/download/124-sari2024.pdf
Hardiansyah, A., & Wibowo, T. (2023). Data-driven inventory forecasting in automotive industry using machine learning. Jurnal Sistem Informasi, 19(4), 355–364. https://jsi.id/download/194-hardiansyah2023.pdf
Andini, N. (2025). Perbandingan performa ARIMA, SARIMA, dan KNN Regression dalam memprediksi penjualan part number di industri manufaktur. Universitas Negeri Jakarta Repository. http://repository.unj.ac.id/id/eprint/56686
Yadav, A., Singh, A., Rao, A., & Raza, M. (2025). A Comparative Study of Artificial Intelligence Algorithms for Health Monitoring of Smart Electric Drivetrain Components. Journal of Smart Industrial Applications and Research, 4(2), 107–119. https://jsiar.com/2025-April/JSIAR-A-25-04107.pdf
Laxmi, M. M., Varshini, K. V. S., & Sandhya, B. (2025). Hybrid Ensemble Learning for Real-Time Predictive Maintenance in Vehicular Engine Health Monitoring Systems. Frontiers in Computing and Robotics, 9(1), 156–167. https://www.macawpublications.com/Journals/index.php/FCR/article/view/156
Khalid, M., & Jasińska, E. (2025). AI-based Energy Storage Systems for Electric Vehicle Performance Optimization. Frontiers in Energy Research, 13, 1626556. https://www.frontiersin.org/journals/energy-research/articles/ 10.3389/fenrg.2025.1626556/pdf
Nguyen, T. H., & Le, Q. T. (2023). Hybrid support vector machine and k-nearest neighbor for demand forecasting of automotive spare parts. International Journal of Industrial Engineering and Operations Management, 5(4), 215–226. https://ijieom.org/wp-content/uploads/2023/07/IJIEOM-54215.pdf
Bhatnagar, A., & Sharma, P. (2025). Hybrid SVM-KNN approach for automotive spare parts demand forecasting. International Journal of Forecasting and Applied Analytics, 12(1), 45–56. https://ijfaa.org/2025/12-1-45.pdf
Li, Y., Chen, Z., & Zhang, H. (2024). Integration of SVM and KNN models for improved demand prediction in automotive aftermarket parts. Journal of Intelligent Manufacturing Systems, 18(2), 89–101. https://jim-systems.com/articles/18-2-li2024.pdf
Ahmed, M., Khalid, S., & Raza, M. (2024). Hybrid machine learning models for spare parts inventory forecasting in automotive industries. International Journal of Supply Chain Management, 13(3), 200–211. https://ojs.excelingtech.co.uk/ index.php/IJSCM/article/view/6200/3184
Wang, X., & Liu, Q. (2024). A comparative study of SVM and KNN ensembles for predicting automotive spare parts demand. International Journal of Advanced Manufacturing Technology, 124, 1123–1137. https://link.springer.com/content/ pdf/10.1007/s00170-023-11541-9.pdf
Gupta, R., & Kumar, S. (2023). Forecasting automotive spare parts requirements using hybrid support vector regression and k-nearest neighbor models. International Journal of Industrial Engineering Computations, 14(2), 157–170. https://ijiec.com/files/gupta2023.pdf
Hoang, N. T., & Tran, M. H. (2023). Machine learning integration for automotive aftermarket demand forecasting: A hybrid SVM-KNN model. Procedia Computer Science, 218, 312–321. https://www.sciencedirect.com/science/ article/pii/S1877050923001125/pdfft
El-Baz, A. H., & El-Shafee, O. M. (2022). A hybrid SVM-KNN framework for automotive spare parts sales forecasting. Journal of Artificial Intelligence and Data Mining, 10(4), 55–65. https://jadam.ai/articles/10-4-elbaz2022.pdf
Rodríguez, J., Martínez, F., & García, P. (2022). Hybrid SVM and KNN techniques for predictive maintenance and spare parts demand. IEEE Access, 10, 82532–82545. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9834523
Kim, S., Park, H., & Lee, J. (2021). Automotive spare parts demand forecasting using combined machine learning approaches. Expert Systems with Applications, 184, 115557. https://www.sciencedirect.com/science/article/abs/pii/S0957417421004515/pdfft
Al-Fuqaha, A., & Qureshi, M. A. (2020). SVM-KNN hybrid model for intelligent inventory control in the automotive industry. International Journal of Production Research, 58(17), 5224–5238. https://www.tandfonline.com/doi/pdf/ 10.1080/00207543.2020.1715502
Fatalifi, A. G. (2025). Pengembangan sistem peramalan permintaan menggunakan algoritma support vector regression untuk optimalisasi safety stock berbasis web. Nusaputra Repository. https://repository.nusaputra.ac.id/id/eprint/ 1441/1/AMERJID%20GHULAMSON%20FATALIFI%20%28Repository%29.pdf
Ardafan, F. H. B. (2024). Aplikasi invoice reader berbasis machine learning sebagai upaya optimasi proses bisnis.
Utama, J. P. (2022). The multi-criteria inventory model of aircraft spare part initial provisioning (A case study of Garuda Indonesia Airbus A330neo entry into service). Institut Teknologi Sepuluh Nopember.
Fatalifi, A. G. (2025). Pengembangan sistem peramalan permintaan menggunakan algoritma support vector regression untuk optimalisasi safety stock berbasis web (Studi kasus: JG). Nusaputra Repository. https://repository.nusaputra.ac.id/ 1441/1/AMERJID%20GHULAMSON%20FATALIFI%20%28Repository%29.pdf
Dewi, B. E. S., Haikal, S., & Sulistyowati, H. S. (2024). Penerapan machine learning menggunakan algoritma random forest untuk prediksi harga mobil bekas. Jurnal Teknologi dan Sistem Komputer, 12(2), 45–55. https://banisalehjurnal.ubs.ac.id/ index.php/tridi/article/download/38/42
Ramadhan, D. I. (2020). Analisis kinerja peramalan dan klasifikasi permintaan part otomotif dengan pendekatan time-series dan data mining. Institut Teknologi Sepuluh Nopember.
Prabowo, T., & Sari, N. (2023). Hybrid SVM and KNN approach for spare parts demand forecasting in the automotive industry. Jurnal Teknologi dan Sistem Komputer, 11(4), 201–210. https://jtsiskom.id/download/1104-prabowo2023.pdf
Nugroho, A. P., & Rahmawati, L. (2022). Integrating support vector machine and k-nearest neighbor algorithms to improve forecasting accuracy for automotive spare parts demand. Jurnal Ilmiah Teknik Industri, 21(2), 145–156. https://jiti.id/files/jiti2022-nugroho.pdf
Setiawan, D., & Lestari, M. (2021). Optimizing demand prediction of automotive spare parts using a hybrid machine learning model. Jurnal Sistem Informasi, 17(1), 33–44. https://jsi.id/download/jsi17-1-setiawan2021.pdf
Wulandari, E., & Pratama, R. (2024). Demand forecasting of motorcycle spare parts using machine learning algorithms. Jurnal Teknologi Informasi dan Ilmu Komputer, 11(2), 145–154. https://jtik.id/download/112-wulandari2024.pdf
Hidayat, M. R., & Fauzi, M. A. (2023). Predictive analytics for spare parts inventory management in the automotive sector using supervised learning. Jurnal Sistem Informasi, 19(3), 211–220. https://jsi.id/download/193-hidayat2023.pdf
Santoso, B., & Wijaya, A. (2022). Forecasting automotive spare parts demand with ensemble machine learning models. Jurnal Ilmu Komputer dan Informasi, 15(1), 57–66. https://jik.id/download/jik151-santoso2022.pdf
Yuliana, S., & Hakim, R. (2024). Optimization of spare parts inventory using machine learning-based demand forecasting. Jurnal Teknologi dan Sistem Komputer, 12(1), 88–97. https://jtsiskom.id/download/121-yuliana2024.pdf
Ramadhani, L. A., & Prasetyo, B. (2023). Hybrid forecasting model for spare parts stock optimization in automotive industry. Jurnal Sistem Informasi, 19(4), 299–308. https://jsi.id/download/194-ramadhani2023.pdf
Kurniawan, D., & Fitria, N. (2021). Application of support vector regression and k-nearest neighbor in spare parts inventory optimization. Jurnal Ilmiah Teknik Industri, 20(3), 177–186. https://jiti.id/files/jiti2021-kurniawan.pdf
Susanto, H., & Lestari, E. (2024). Automotive sales demand forecasting using hybrid SVM and KNN models. Jurnal Teknologi Informasi dan Ilmu Komputer, 12(2), 122–131. https://jtik.id/download/122-susanto2024.pdf
Maulana, R., & Saputra, A. (2023). Comparative analysis of SVM and KNN for demand prediction in the automotive industry. Jurnal Sistem Informasi, 19(2), 167–176. https://jsi.id/download/192-maulana2023.pdf
Anggraini, D., & Hidayat, T. (2021). Improving accuracy of vehicle demand forecasting using machine learning hybrid approach. Jurnal Ilmu Komputer dan Informasi, 14(4), 291–300. https://jik.id/download/jik144-anggraini2021.pdf
Putri, V. A., & Gunawan, F. (2024). Demand prediction for automotive supply chain using hybrid machine learning models. Jurnal Teknologi dan Sistem Komputer, 12(3), 201–210. https://jtsiskom.id/download/123-putri2024.pdf
Akbar, M. R., & Suryani, T. (2023). Machine learning-based forecasting for supply chain demand in the automotive sector. Jurnal Sistem Informasi, 19(1), 45–54. https://jsi.id/download/191-akbar2023.pdf
Dewantara, H., & Prasetia, A. (2021). Application of hybrid SVM-KNN models in automotive supply chain demand forecasting. Jurnal Ilmiah Teknik Industri, 20(1), 33–42. https://jiti.id/files/jiti2021-dewantara.pdf
Kusuma, V., & Budhidharma, V. (2025). Inventory management optimization in automotive spare parts industry using machine learning approach. Jurnal Ilmiah Multidisiplin, 5(1), 88–99. https://ejournal.seaninstitute.or.id/ index.php/esaprom/article/download/6537/5072
Sari, P. A., & Nugraha, R. (2024). Predictive analytics for automotive spare parts inventory using hybrid SVM and KNN. Jurnal Teknologi dan Sistem Komputer, 12(4), 277–286. https://jtsiskom.id/download/124-sari2024.pdf
Hardiansyah, A., & Wibowo, T. (2023). Data-driven inventory forecasting in automotive industry using machine learning. Jurnal Sistem Informasi, 19(4), 355–364. https://jsi.id/download/194-hardiansyah2023.pdf
Andini, N. (2025). Perbandingan performa ARIMA, SARIMA, dan KNN Regression dalam memprediksi penjualan part number di industri manufaktur. Universitas Negeri Jakarta Repository. http://repository.unj.ac.id/id/eprint/56686
Yadav, A., Singh, A., Rao, A., & Raza, M. (2025). A Comparative Study of Artificial Intelligence Algorithms for Health Monitoring of Smart Electric Drivetrain Components. Journal of Smart Industrial Applications and Research, 4(2), 107–119. https://jsiar.com/2025-April/JSIAR-A-25-04107.pdf
Laxmi, M. M., Varshini, K. V. S., & Sandhya, B. (2025). Hybrid Ensemble Learning for Real-Time Predictive Maintenance in Vehicular Engine Health Monitoring Systems. Frontiers in Computing and Robotics, 9(1), 156–167. https://www.macawpublications.com/Journals/index.php/FCR/article/view/156
Khalid, M., & Jasińska, E. (2025). AI-based Energy Storage Systems for Electric Vehicle Performance Optimization. Frontiers in Energy Research, 13, 1626556. https://www.frontiersin.org/journals/energy-research/articles/ 10.3389/fenrg.2025.1626556/pdf
Nguyen, T. H., & Le, Q. T. (2023). Hybrid support vector machine and k-nearest neighbor for demand forecasting of automotive spare parts. International Journal of Industrial Engineering and Operations Management, 5(4), 215–226. https://ijieom.org/wp-content/uploads/2023/07/IJIEOM-54215.pdf
Bhatnagar, A., & Sharma, P. (2025). Hybrid SVM-KNN approach for automotive spare parts demand forecasting. International Journal of Forecasting and Applied Analytics, 12(1), 45–56. https://ijfaa.org/2025/12-1-45.pdf
Li, Y., Chen, Z., & Zhang, H. (2024). Integration of SVM and KNN models for improved demand prediction in automotive aftermarket parts. Journal of Intelligent Manufacturing Systems, 18(2), 89–101. https://jim-systems.com/articles/18-2-li2024.pdf
Ahmed, M., Khalid, S., & Raza, M. (2024). Hybrid machine learning models for spare parts inventory forecasting in automotive industries. International Journal of Supply Chain Management, 13(3), 200–211. https://ojs.excelingtech.co.uk/ index.php/IJSCM/article/view/6200/3184
Wang, X., & Liu, Q. (2024). A comparative study of SVM and KNN ensembles for predicting automotive spare parts demand. International Journal of Advanced Manufacturing Technology, 124, 1123–1137. https://link.springer.com/content/ pdf/10.1007/s00170-023-11541-9.pdf
Gupta, R., & Kumar, S. (2023). Forecasting automotive spare parts requirements using hybrid support vector regression and k-nearest neighbor models. International Journal of Industrial Engineering Computations, 14(2), 157–170. https://ijiec.com/files/gupta2023.pdf
Hoang, N. T., & Tran, M. H. (2023). Machine learning integration for automotive aftermarket demand forecasting: A hybrid SVM-KNN model. Procedia Computer Science, 218, 312–321. https://www.sciencedirect.com/science/ article/pii/S1877050923001125/pdfft
El-Baz, A. H., & El-Shafee, O. M. (2022). A hybrid SVM-KNN framework for automotive spare parts sales forecasting. Journal of Artificial Intelligence and Data Mining, 10(4), 55–65. https://jadam.ai/articles/10-4-elbaz2022.pdf
Rodríguez, J., Martínez, F., & García, P. (2022). Hybrid SVM and KNN techniques for predictive maintenance and spare parts demand. IEEE Access, 10, 82532–82545. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9834523
Kim, S., Park, H., & Lee, J. (2021). Automotive spare parts demand forecasting using combined machine learning approaches. Expert Systems with Applications, 184, 115557. https://www.sciencedirect.com/science/article/abs/pii/S0957417421004515/pdfft
Al-Fuqaha, A., & Qureshi, M. A. (2020). SVM-KNN hybrid model for intelligent inventory control in the automotive industry. International Journal of Production Research, 58(17), 5224–5238. https://www.tandfonline.com/doi/pdf/ 10.1080/00207543.2020.1715502