Prediksi Pengiriman Bermasalah PT Lion Express Palembang: Perbandingan Tiga Model Klasifikasi
DOI:
https://doi.org/10.63822/q3ntna05Keywords:
klasifikasi, machine learning, pengiriman bermasalah, data tidak seimbang, SMOTENC, Random ForestAbstract
Pengiriman bermasalah (delivery exception) menambah biaya penanganan dan menurunkan kepuasan pelanggan, tetapi proporsinya yang kecil menyulitkan identifikasi manual. Penelitian ini membandingkan Logistic Regression, Decision Tree, dan Random Forest untuk memprediksi pengiriman yang belum berstatus bukti pengiriman (non-POD) pada PT Lion Express Cabang Palembang berdasarkan area tujuan, berat tertagih, siklus, dan hari pengiriman, serta mengevaluasi penanganan ketidakseimbangan kelas. Data terdiri atas 32.475 transaksi Agustus 2026 dengan proporsi kelas bermasalah 8,52%. Data dibagi per nomor resi untuk mencegah kebocoran, hyperparameter dioptimasi dengan validasi silang 5-lipatan berkriteria PR-AUC, dan ketidakseimbangan kelas ditangani dengan SMOTENC. Pada data uji, Random Forest memperoleh ROC-AUC 0,745 (CI 95%: 0,719–0,770) dan PR-AUC 0,370 (4,3 kali nilai acak 0,085); selisihnya dengan Decision Tree tidak signifikan. SMOTENC tidak meningkatkan ROC-AUC maupun PR-AUC, tetapi menggeser titik operasi, sedangkan tuning menaikkan PR-AUC sekitar 0,03. Area tujuan merupakan fitur terpenting, diikuti hari, siklus, dan berat tertagih. Pada validasi temporal, ROC-AUC turun menjadi 0,666 dan PR-AUC 0,143, sehingga model lebih sesuai untuk prioritisasi pengawasan daripada keputusan otomatis.
References
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357. https://doi.org/10.1613/jair.953
Davis, J., & Goadrich, M. (2006). The relationship between Precision-Recall and ROC curves. In Proceedings of the 23rd International Conference on Machine Learning (ICML) (pp. 233–240). ACM. https://doi.org/10.1145/1143844.1143874
de Leeuw, J., Bukhsh, Z., & Zhang, Y. (2023). Parcel loss prediction in last-mile delivery: Deep and non-deep approaches with insights from Explainable AI [Preprint]. arXiv. https://arxiv.org/abs/2310.16602
Elor, Y., & Averbuch-Elor, H. (2022). To SMOTE, or not to SMOTE? [Preprint]. arXiv. https://arxiv.org/abs/2201.08528
Erlin, E., Desnelita, Y., Nasution, N., Suryati, L., & Zoromi, F. (2022). Dampak SMOTE terhadap kinerja Random Forest classifier berdasarkan data tidak seimbang. MATRIK: Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer, 21(3), 677–690. https://doi.org/10.30812/matrik.v21i3.1726
Fawcett, T. (2006). An introduction to ROC analysis. Pattern Recognition Letters, 27(8), 861–874. https://doi.org/10.1016/j.patrec.2005.10.010
Hayati, E., Nurjanah, F., & Kurnia Laili, F. (2024). Unbalanced timeliness of financial reporting data classification using Random Forest with SMOTE. Inform: Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi, 9(2), 132–136. https://doi.org/10.25139/inform.v9i2.8327
Lemaître, G., Nogueira, F., & Aridas, C. K. (2017). Imbalanced-learn: A Python toolbox to tackle the curse of imbalanced datasets in machine learning. Journal of Machine Learning Research, 18(17), 1–5. https://jmlr.org/papers/v18/16-365.html
Lim, S. F. W. T., Wang, Q., & Webster, S. (2023). Do it right the first time: Vehicle routing with home delivery attempt predictors. Production and Operations Management, 32(4), 1262–1284. https://doi.org/10.1111/poms.13926
Micci-Barreca, D. (2001). A preprocessing scheme for high-cardinality categorical attributes in classification and prediction problems. ACM SIGKDD Explorations Newsletter, 3(1), 27–32. https://doi.org/10.1145/507533.507538
Pambudi, H. K., Kusuma, P. G. A., Yulianti, F., & Julian, K. A. (2020). Prediksi status pengiriman barang menggunakan metode machine learning. Jurnal Ilmiah Teknologi Informasi Terapan, 6(2), 100–109. https://doi.org/10.33197/jitter.vol6.iss2.2020.396
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., & Duchesnay, E. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830. https://jmlr.org/papers/v12/pedregosa11a.html
Richardson, E., Trevizani, R., Greenbaum, J. A., Carter, H., Nielsen, M., & Peters, B. (2024). The receiver operating characteristic curve accurately assesses imbalanced datasets. Patterns, 5(6), 100994. https://doi.org/10.1016/j.patter.2024.100994
Saito, T., & Rehmsmeier, M. (2015). The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLOS ONE, 10(3), e0118432. https://doi.org/10.1371/journal.pone.0118432
Strobl, C., Boulesteix, A.-L., Zeileis, A., & Hothorn, T. (2007). Bias in random forest variable importance measures: Illustrations, sources and a solution. BMC Bioinformatics, 8, Article 25. https://doi.org/10.1186/1471-2105-8-25
van den Goorbergh, R., van Smeden, M., Timmerman, D., & Van Calster, B. (2022). The harm of class imbalance corrections for risk prediction models: Illustration and simulation using logistic regression. Journal of the American Medical Informatics Association, 29(9), 1525–1534. https://doi.org/10.1093/jamia/ocac093
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Muktiono Muktiono, Rusmala Santi (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
