Analysis of Wagon Repair Data Using SOM and K-Means Clustering Algorithms
Yıl 2025,
Sayı: 21, 168 - 177, 31.01.2025
Ender Gunher
,
Mehmet Fidan
,
Ömür Akbayır
Öz
This study aims to investigate the comparative performances of Self Organizing Maps (SOM) and K-Means algorithms for the optimization of railway maintenance processes in Turkey. For this purpose, the performances of SOM and K-Means clustering algorithms used in the determination of the railcar repair location (TTY) were investigated. The dataset used in the study was obtained from railcar failure records in Turkey and includes attributes such as Reason for Repair (TTN), Component Name (KA) and Railcar Type (VT). SOM and K-Means algorithms were applied on the dataset where the attributes of Reason for Repair (TTN), Component Name (KA) and Wagon Type (VT) were used. SOM enables the visualization of high-dimensional data on a two-dimensional map and enables the collection of data with similar characteristics in the same cluster. The K-Means algorithm assigns data points to a certain number of cluster centers and collects the data points closest to these centers in the same cluster. The analysis results show that SOM and K-Means algorithms are effective in optimizing the wagon maintenance processes. Using these methods together will allow the management of wagon maintenance processes in a more efficient and systematic way. Correct fault detection and directing to the appropriate workshops will contribute to the acceleration of maintenance processes and the reduction of costs.
Kaynakça
- [1] F. Feuillet, vd. "Psikotrop ilaçların tüketimi ve istatistiksel metodlar," Journal of Health Studies, vol. 45, no. 3, pp. 123-135, 2012
- [2] T.S. Madhulatha, "Kümeleme algoritmaları ve veri madenciliği," Data Mining Journal, vol. 34, no. 2, pp. 67-89, 2012
- [3] W. Zhang, Z. Wang, Z. Jia, H. Wang, “Optimization model for collaborative overhaul workshop scheduling problem of multiple EMUs,” in 2021 IEEE 23rd Int Conf on High Performance Computing & Communications; 7th Int Conf on Data Science & Systems; 19th Int Conf on Smart City; 7th Int Conf on Dependability in Sensor Cloud & Big Data Systems & Application, 2021, doi: 10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00181
- [4] Z. Wang, Q. Deng, L. Zhang, H. Li, F. Li, “Joint optimization of integrated mixed maintenance and distributed two-stage hybrid flow-shop production for multi-site maintenance requirements,” Expert Systems with Applications, 215, 119422, 2023
- [5] M. Rodoplu, S. Dauzère-Pérès, P. Vialletelle, “Integrated planning of maintenance operations and workload allocation,” International Journal of Production Research, vol. 61, no. 23, pp. 8291-8308, 2023
- [6] Y. Zhang, C. Li, X. Su, R. Cui, B. Wan, “A baseline-reactive scheduling method for carrier-based aircraft maintenance tasks,” Complex & Intelligent Systems, vol. 9, no. 1, pp. 367-397, 2023
- [7] W. T. Lin, Y. C. Wu, J. S. Zheng, M. Y. Chen, “Analysis by data mining in the emergency medicine triage database at a Taiwanese regional hospital,” Expert Systems with Applications, vol. 38, no. 9, pp. 11078-11084, 2011.
- [8] K. R. Kashwan, C. M. Velu, “Customer segmentation using clustering and data mining techniques,” International Journal of Computer Theory and Engineering, vol. 5, no. 6, pp. 856, 2013.
- [9] L. Pinciroli, P. Baraldi, E. Zio, “Maintenance optimization in industry 4.0,” Reliability Engineering & System Safety, vol. 234, pp. 109204, 2023.
- [10] I. Al-Nader, A. Lasebae, R. Raheem, “A novel scheduling algorithm for improved performance of multi-objective safety-critical WSN using spatial self-organizing feature map,” Electronics, vol. 13, no. 1, 19, 2023
- [11] S. Sivaraju, C. Kumar, “Energy enhancement of WSN with deep learning based SOM scheduling algorithm,” Journal of Information Technology and Digital World, vol. 4, no. 3, pp. 238-249, 2022
- [12] S. Shadroo, A.M. Rahmani, A. Rezaee, “The two-phase scheduling based on deep learning in the Internet of Things,” Computer Networks, 185, 107684, 2021
- [13] A. Singh, G.S. Aujla, R.S. Bali, P.K. Chahal, M. Singh, “A self organised workload classification and scheduling approach in IoT-edge-cloud ecosystem,” 2020 IEEE 92nd Vehicular Technology Conference (VTC2020-Fall), pp. 1-5, 2020
- [14] S.M. Mostafa, H. Amano, “Dynamic round robin CPU scheduling algorithm based on K-means clustering technique,” Applied Sciences, vol. 10, no. 15, 5134, 2020
- [15] M. Belhor, A. El-Amraoui, A. Jemai, F. Delmotte, “Multi-objective evolutionary approach based on K-means clustering for home health care routing and scheduling problem,” Expert Systems with Applications, 213, 119035, 2023
- [16] I. Ullah, H.Y. Youn, “Task classification and scheduling based on K-means clustering for edge computing,” Wireless Personal Communications, vol. 113, no. 4, pp. 2611-2624, 2020
- [17] J. Wang, “Maintenance scheduling at high-speed train depots: An optimization approach, ” Reliability Engineering & System Safety, vol. 243, pp.109809, 2024.
SOM ve K-Ortalama Kümeleme Algoritmaları Kullanarak Vagon Tamire Tutma Verilerinin İncelenmesi
Yıl 2025,
Sayı: 21, 168 - 177, 31.01.2025
Ender Gunher
,
Mehmet Fidan
,
Ömür Akbayır
Öz
Bu çalışma, Türkiye demiryolu bakım süreçlerinin optimizasyonu için Kendini Organize Eden Haritalar (SOM) ve K-Ortalama algoritmalarının karşılaştırmalı performanslarını incelemeyi amaçlamaktadır. Bu amaçla vagon tamire tutulma yeri (TTY) tespitinde kullanılan SOM ve K-Ortalama kümeleme algoritmalarının performansları incelenmiştir. Çalışmada kullanılan veri seti, Türkiye'deki vagon arıza kayıtlarından elde edilmiş olup, Tamire Tutulma Nedeni (TTN), Komponent Adı (KA) ve Vagon Tipi (VT) gibi öznitelikleri içermektedir. SOM ve K-Ortalama algoritmaları, Tamire Tutulma Nedeni (TTN), Komponent Adı (KA) ve Vagon Tipi (VT) özniteliklerinin kullanıldığı veri seti üzerinde uygulanmıştır. SOM, yüksek boyutlu verilerin iki boyutlu bir harita üzerinde görselleştirilmesini sağlayarak benzer özelliklere sahip verilerin aynı kümede toplanmasına imkân tanır. K-Ortalama algoritması ise veri noktalarını belirli sayıda küme merkezine atayarak bu merkezlere en yakın veri noktalarını aynı kümede toplar. Analiz sonuçları, SOM ve K-Ortalama algoritmalarının vagon bakım süreçlerini optimize etme açısından etkili olduğunu göstermektedir. Bu yöntemlerin birlikte kullanılması, vagon bakım süreçlerinin daha verimli ve sistematik bir şekilde yönetilmesine olanak tanıyacaktır. Doğru arıza tespiti ve uygun atölyelere yönlendirme, bakım süreçlerinin hızlanmasına ve maliyetlerin düşürülmesine katkı sağlayacaktır.
Destekleyen Kurum
TCDD Taşımacılık AŞ
Teşekkür
Bu çalışma TCDD Taşımacılık AŞ’nin 04.10.2021 tarih ve E-30614766-204.02.99-128179 sayılı onayı ile yapılmıştır.
Kaynakça
- [1] F. Feuillet, vd. "Psikotrop ilaçların tüketimi ve istatistiksel metodlar," Journal of Health Studies, vol. 45, no. 3, pp. 123-135, 2012
- [2] T.S. Madhulatha, "Kümeleme algoritmaları ve veri madenciliği," Data Mining Journal, vol. 34, no. 2, pp. 67-89, 2012
- [3] W. Zhang, Z. Wang, Z. Jia, H. Wang, “Optimization model for collaborative overhaul workshop scheduling problem of multiple EMUs,” in 2021 IEEE 23rd Int Conf on High Performance Computing & Communications; 7th Int Conf on Data Science & Systems; 19th Int Conf on Smart City; 7th Int Conf on Dependability in Sensor Cloud & Big Data Systems & Application, 2021, doi: 10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00181
- [4] Z. Wang, Q. Deng, L. Zhang, H. Li, F. Li, “Joint optimization of integrated mixed maintenance and distributed two-stage hybrid flow-shop production for multi-site maintenance requirements,” Expert Systems with Applications, 215, 119422, 2023
- [5] M. Rodoplu, S. Dauzère-Pérès, P. Vialletelle, “Integrated planning of maintenance operations and workload allocation,” International Journal of Production Research, vol. 61, no. 23, pp. 8291-8308, 2023
- [6] Y. Zhang, C. Li, X. Su, R. Cui, B. Wan, “A baseline-reactive scheduling method for carrier-based aircraft maintenance tasks,” Complex & Intelligent Systems, vol. 9, no. 1, pp. 367-397, 2023
- [7] W. T. Lin, Y. C. Wu, J. S. Zheng, M. Y. Chen, “Analysis by data mining in the emergency medicine triage database at a Taiwanese regional hospital,” Expert Systems with Applications, vol. 38, no. 9, pp. 11078-11084, 2011.
- [8] K. R. Kashwan, C. M. Velu, “Customer segmentation using clustering and data mining techniques,” International Journal of Computer Theory and Engineering, vol. 5, no. 6, pp. 856, 2013.
- [9] L. Pinciroli, P. Baraldi, E. Zio, “Maintenance optimization in industry 4.0,” Reliability Engineering & System Safety, vol. 234, pp. 109204, 2023.
- [10] I. Al-Nader, A. Lasebae, R. Raheem, “A novel scheduling algorithm for improved performance of multi-objective safety-critical WSN using spatial self-organizing feature map,” Electronics, vol. 13, no. 1, 19, 2023
- [11] S. Sivaraju, C. Kumar, “Energy enhancement of WSN with deep learning based SOM scheduling algorithm,” Journal of Information Technology and Digital World, vol. 4, no. 3, pp. 238-249, 2022
- [12] S. Shadroo, A.M. Rahmani, A. Rezaee, “The two-phase scheduling based on deep learning in the Internet of Things,” Computer Networks, 185, 107684, 2021
- [13] A. Singh, G.S. Aujla, R.S. Bali, P.K. Chahal, M. Singh, “A self organised workload classification and scheduling approach in IoT-edge-cloud ecosystem,” 2020 IEEE 92nd Vehicular Technology Conference (VTC2020-Fall), pp. 1-5, 2020
- [14] S.M. Mostafa, H. Amano, “Dynamic round robin CPU scheduling algorithm based on K-means clustering technique,” Applied Sciences, vol. 10, no. 15, 5134, 2020
- [15] M. Belhor, A. El-Amraoui, A. Jemai, F. Delmotte, “Multi-objective evolutionary approach based on K-means clustering for home health care routing and scheduling problem,” Expert Systems with Applications, 213, 119035, 2023
- [16] I. Ullah, H.Y. Youn, “Task classification and scheduling based on K-means clustering for edge computing,” Wireless Personal Communications, vol. 113, no. 4, pp. 2611-2624, 2020
- [17] J. Wang, “Maintenance scheduling at high-speed train depots: An optimization approach, ” Reliability Engineering & System Safety, vol. 243, pp.109809, 2024.