Kocaeli Akçaray Tramvay Sisteminin Bakım ve Onarım Modelinin Veri Madenciliği ile İyileştirilmesi

Öz

Ulaşım birçok insanın günlük rutinleri arasında yer almakta ve insan hayatının büyük bir kısmı ulaşımda geçmektedir. Ulaşımın bu kadar önem kazandığı bir süreçte konforlu, güvenli ve hızlı ulaşım için araçların bakım ve onarımlarının düzenli olarak yapılması önem kazanmaktadır. Bu çalışmada Kocaeli Büyükşehir Belediyesi tramvay işletmeciliğinin bakım ve onarım sürelerinin iyileştirilmesi amaçlanmıştır. Bu nedenle Veri Madenciliği algoritmalarından K-NN (K-en yakın komşular) sınıflandırması, Karar Ağaçları ve Derin Öğrenme algoritmaları uygulanarak Bakım Onarım Veri Modeli tasarlanmıştır. Çalışmada kullanılan üç algoritmadan elde edilen sonuçlar karşılaştırılmış ve en iyi bakım süresi performansı K-NN algoritması ile elde edilmiştir. Böylece tramvayların garanti süreleri uzatılmakta ve arızalar arasındaki ortalama süre ve bakım çalışma periyotlarının uzunluğuna bakılarak tramvayların maksimum performansla çalışması sağlanmaktadır.

Anahtar Kelimeler

• Veri Madenciliği
• K-NN
• Karar Ağaçları
• Derin Öğrenme
• Tramvay Bakım ve Onarımı

Improving the Maintenance and Repair Model of Kocaeli Akçaray Tram System with Data Mining

Öz

Transportation is among the daily routines of many people and a large part of human life is spent in transportation. In a process where transportation has gained so much importance, it is important to regularly perform maintenance and repair of vehicles for comfortable, safe and fast transportation. In this study, it is aimed to improve the maintenance and repair times of Kocaeli Metropolitan Municipality tram operation. Therefore, Maintenance Repair Data Model is designed by applying K-NN (K-nearest neighbors) classification, Decision Trees and Deep Learning algorithms from Data Mining algorithms.Thus, the warranty periods of the trams are extended and the trams operate with maximum performance by looking at the mean time between failures and the length of maintenance work periods.Quantitative findings revealed that the K-NN algorithm achieved the highest performance with an accuracy rate of 87.6%, outperforming the Decision Tree 85.95% and Deep Learning 70.66% models. Moreover, the K-NN model demonstrated the most balanced classification performance, with a precision of 0.716, recall of 0.637, and an F1-score of 0.652. In contrast, the Deep Learning algorithm exhibited lower performance, with an F1-score of only 0.355, indicating its limited effectiveness when applied to structured maintenance data. These results suggest that in cases where the dataset is structured and relatively small in scale, simpler non-parametric models such as K-NN may offer more efficient and practical solutions for predicting maintenance durations.

Anahtar Kelimeler

• Data Mining
• K-NN
• Decision Trees
• Deep Learning
• Tram Maintenance and Reapir

Kaynakça

1. [1] Aydın F. M., 2021. Comparison of traditional and rubber tyred tram (translohr) systems and investigation of applicability in Erzincan, Msc. Thesis, Erzincan Binali Yıldırım University, Institute of Science.
2. [2] Mahale Y., Kolhar S. and More A. S., 2025. A comprehensive review on artificial intelligence driven predictive maintenance in vehicles: technologies, challenges and future research directions, Discover Applied Sciences,7(4), 243.
3. [3] Le-Nguyen M. H., Turgis F., Fayemi P.-E., Bifet, A., 2023. Real-time learning for real-time data: Online machine learning for predictive maintenance of railway systems, Transportation Research Procedia, 72, pp.171–178.
4. [4] Ahac M., Lakusic S., 2017. Track Gauge Degradation Modelling on Small Urban Rail Networks: Zagreb Tram System Case Study, Urban Transport Systems.
5. [5] Arvidsson N., Browne M., 2013. A review of the success and failure of tram systems to carry urban freight: the implications for a low emission intermodal solution using electric vehicles on trams, European Transport-TransportiEuropei, (54), pp. 1-18.
6. [6] Liden T., 2015. Railway infrastructure maintenance –a survey of planning problems and conducted research, Transportation Research Procedia, 10, pp. 574-583.
7. [7] Ahac M., Lakusic S., 2015. Tram track maintenance planning by gauge degradationmodelling, Transport, 30, pp. 430-436.
8. [8] Yousefikia M., Moridpour S., Setunge S., and Mazloumi E., 2014. Modeling degradation of tracks for maintenance planning on a tram line, Journal of Traffic and Logistics Engineering, 2(2), pp. 86–91.

9. [9] MajidiParast S., Monemi R. N., and Gelareh S., 2025. A graph convolutional network for optimal intelligent predictive maintenance of railway tracks, Decision Analytics Journal, 14, 100542.
10. [10] Bukhsh Z. A., Saeed A., Stipanovic I. and Doree A. G., 2019. Predictive maintenance using tree-based classification techniques: A case of railway switches, Transportation Research Part C: Emerging Technologies, 101, 35-54.
11. [11] Nair V., Premalatha M., and Braveen, M., 2024. Enhancing metro rail efficiency: A predictive maintenance approach leveraging machine learning and deep learning technologies.
12. [12] Kowalski M., Magott J., Nowakowski T., Werbińska-Wojciechowska S., 2014.Exact and approximation methods for dependability assessment of tram systems with time window, European Journal of Operational Research, 235(3), pp. 671-686.
13. [13] Carrese S., Ottone G., 2006. A model for the management of a tram fleet, European Journal of Operational Research, 175(3), pp. 1628-1651.
14. [14] Gürbüz F., Turna F., 2018. Rule extraction for tram faults via data mining for safe transportation, Transportation research part A: policy and practice, 116, pp. 568-579.
15. [15] Gökgöz K. B., 2015. Medical equipment maintenance decision model with data mining, Msc. Thesis, Başkent University, Institute of Social Science.
16. [16] Kiefer A., Schilde M., Doerner K. F., 2018. Scheduling of maintenance work of a large-scale tramway network, European Journal of Operational Research, 270(3), pp. 1158-1170.
17. [17] Turna F., 2011. Rule extraction for tram faults via data mining, Msc. Thesis, Erciyes University, Institute of Science.
18. [18] YadavA. K., Malik H., Chandel S. S., 2015. Application of rapid miner in ANN based prediction of solar radiation for assessment of solar energy resource potential of 76 sites in Northwestern India, Renewable and Sustainable Energy Reviews, 52, pp. 1093-1106.
19. [19] Sudarsono B. G., Leo M. I., Santoso A., Hendrawan F., 2021. Analisis Data Mining Data Netflix Menggunakan Aplikasi Rapid Miner, JBASE-Journal of Business and Audit Information Systems, 4(1).
20. [20] Balta S., 1998. Analysis of electric consumption data by data mining methods and determining the right tariff, Msc. Thesis, Sakarya University, Institute of Science.
21. [21] Yıldırım M. E., 2024. Analysis of maintenance and repair processes of Kocaeli Akçaray transportation system using data mining, Msc. Thesis, Kocaeli University, Institute of Science.
22. [22] Esteban A., Zafra A., and Ventura, S., 2022. Data mining in predictive maintenance systems: A taxonomy and systematic review. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 12(5), e1471.
23. [23] Dorfman R., 1979. A formula for the Gini coefficient. The review of economics and ststistics, pp. 146-149,
24. [24] Zhang S., Li X., Zong M., Zhu X., Cheng D., 2017. Learning k for knn classification, ACM Transactions on Intelligent Systems and Technology, 8(3), pp. 1-19.
25. [25] Lei C., 2021. Deep Learning Methods and Applications.In: Deep Learning and Practice with MindSpore, Cognitive Intelligence and Robotics, Springer.
26. [26] Ahmad J., Muhammad K., Lloret J. and Baik S. W., 2018. Efficient Conversion of Deep Features to Compact Binary Codes Using Fourier Decomposition for Multimedia Big Data, IEEE Transactions on Industrial Informatics, 14(7), pp. 3205-3215.
27. [27] Chen, L, 2021.Deep learning and practice with mindspore. Springer Nature.