PSO-OPTIMIZED XGBOOST FOR MAIL DELIVERY DELAY PREDICTION AND LOGISTICS SLA COMPLIANCE
DOI:
https://doi.org/10.33480/jitk.v12i1.8282Keywords:
Delivery Delay Risk Prediction, Particle Swarm Optimization, Postal Letter Logistic, Service Level Agreement (SLA), XGBoostAbstract
The Indonesian postal logistics sector continues to face challenges in maintaining letter delivery timeliness and Service Level Agreement (SLA) compliance under dynamic operational conditions. Conventional predictive approaches often rely on static representations and are limited in capturing temporal risk patterns in mail delivery processes. This study proposes a data-driven machine learning framework to predict delivery delay risk in postal letter services by integrating temporal feature engineering, class imbalance handling, and metaheuristic-based hyperparameter optimization. The framework applies the Synthetic Minority Over-sampling Technique (SMOTE) and evaluates multiple classification models using stratified cross-validation. Among the evaluated algorithms, XGBoost optimized using Particle Swarm Optimization (PSO) demonstrates the strongest predictive performance. The PSO-optimized configuration (n = 96, lr = 0.1718, d = 3) achieves an accuracy of 0.6605, ROC–AUC of 0.6883, and F1-score of 0.6059, indicating improved class-sensitive prediction. Model interpretability is examined using Mean Decrease in Impurity (MDI), which identifies posting day as the dominant contributor to delivery delays, followed by SLA commitment and intra-day posting patterns. The final framework generates probabilistic risk scores from 0 to 100 percent, with the highest observed value reaching 99.44, enabling early warning and prescriptive operational interventions for potential SLA violations. These results indicate that the proposed PSO–XGBoost framework supports proactive logistics risk management. However, this study is limited to historical data from a single postal operational environment and does not incorporate external factors such as weather, traffic, or regional delivery variations.
Downloads
References
[1] C. Peoples, A. Moore, and N. Georgalas, “Service Level Agreement (SLA) Chains Supported by Cloud in a Complex Port Ecosystem with Competing Stakeholder Goals,” EAI Endorsed Trans. Cloud Syst., p. 174394, 2022, doi: 10.4108/eai.26-7-2022.174394.
[2] O. A. Alamri, “Analysing service level agreements with multiple customers.,” RMIT University, 2024.
[3] R. E. R. Shawon, “Enhancing supply chain resilience across US regions using machine learning and logistics performance analytics,” Int. J. Appl. Math., vol. 38, no. 4s, pp. 182–205, 2025.
[4] P. Mahajan, S. Uddin, F. Hajati, and M. A. Moni, “Ensemble learning for disease prediction: A review,” in Healthcare, 2023, p. 1808, doi: 10.3390/healthcare11121808.
[5] W. Zhou and Z. Xie, “Enhancing Sealing Performance Predictions: A Comprehensive Study of XGBoost and Polynomial Regression Models with Advanced Optimization Techniques,” Materials (Basel)., vol. 18, no. 10, p. 2392, 2025, doi: 10.3390/materials18102392.
[6] R. Garine and R. K. Chakrabortty, “A deep learning and policy optimization approach for supply chain order classification,” Supply Chain Anal., p. 100166, 2025, doi: 10.1016/j.sca.2025.100166.
[7] A. G. Gad, “Particle swarm optimization algorithm and its applications: A systematic review.,” Arch. Comput. methods Eng., vol. 29, no. 5, 2022, doi: 10.1007/s11831-021-09694.
[8] T. M. Shami, A. A. El-Saleh, M. Alswaitti, Q. Al-Tashi, M. A. Summakieh, and S. Mirjalili, “Particle swarm optimization: A comprehensive survey,” IEEE Access, vol. 10, pp. 10031–10061, 2022, doi: 10.1109/ACCESS.2022.3142817.
[9] A. M. Musolf, E. R. Holzinger, J. D. Malley, and J. E. Bailey-Wilson, “What makes a good prediction? Feature importance and beginning to open the black box of machine learning in genetics,” Hum. Genet., vol. 141, no. 9, pp. 1515–1528, 2022, doi: 10.1007/s00439-021-02402-z.
[10] H. Wang, Q. Liang, J. T. Hancock, and T. M. Khoshgoftaar, “Feature selection strategies: a comparative analysis of SHAP-value and importance-based methods,” J. Big Data, vol. 11, no. 1, p. 44, 2024, doi: 10.1186/s40537-024-00905-w.
[11] S. Fei, L. Li, Z. Han, Z. Chen, and Y. Xiao, “Combining novel feature selection strategy and hyperspectral vegetation indices to predict crop yield,” Plant Methods, vol. 18, no. 1, p. 119, 2022, doi: 10.1186/s13007-022-00949-0.
[12] M. A. Khan, A. Kanwal, S. Abbas, F. Khan, and T. Whangbo, “Intelligent Model for Predicting the Quality of Services Violation.,” Comput. Mater. & Contin., vol. 71, no. 2, 2022, doi: 10.32604/cmc.2022.023480.
[13] M. K. Banjanin, M. Stojčić, DJordje Popović, D. Andjelković, G. Jauševac, and M. Husić, “Classification Machine Learning Models for Enhancing the Sustainability of Postal System Modules Within the Smart Transportation Concept,” Sustainability, vol. 17, no. 19, p. 8718, 2025, doi: 10.3390/su17198718.
[14] A. Alabrah, “An improved CCF detector to handle the problem of class imbalance with outlier normalization using IQR method,” Sensors, vol. 23, no. 9, p. 4406, 2023, doi: 10.3390/s23094406.
[15] S. Datta, C. Ghosh, and J. P. Choudhury, “Classification of imbalanced datasets utilizing the synthetic minority oversampling method in conjunction with several machine learning techniques,” Iran J. Comput. Sci., vol. 8, no. 1, pp. 51–68, 2025, doi: 10.1007/s42044-024-00207-7.
[16] M. M. H. Sizan et al., “AI-Enhanced Stock Market Prediction: Evaluating Machine Learning Models for Financial Forecasting in the USA,” J. Bus. Manag. Stud., vol. 5, no. 4, pp. 152–166, 2023, doi: 10.46828/jbms.v5i4.8609.
[17] S. F. Pane, M. D. Sulistiyo, A. A. Gozali, and A. Adiwijaya, “PSO-Enhanced ensemble techniques for pandemic prediction and feature importance analysis,” Int. J. Adv. Intell. Informatics, vol. 11, no. 4, pp. 653–666, 2025, doi: 10.26555/ijain.v11i4.2091.
[18] A.-H. M. Emara, G. Atteia, and J. H. Alkhateeb, “Fine Tuning Hyperparameters of Deep Learning Models Using Metaheuristic Accelerated Particle Swarm Optimization Algorithm,” IEEE Access, 2025, doi: 10.1109/ACCESS.2025.11087552.
[19] Y.-X. He, S.-H. Lyu, and Y. Jiang, “Interpreting deep forest through feature contribution and mdi feature importance,” ACM Trans. Knowl. Discov. Data, 2024, doi: 10.1145/3641108.
[20] A. Agarwal, A. M. Kenney, Y. S. Tan, T. M. Tang, and B. Yu, “MDI+: A flexible random forest-based feature importance framework,” arXiv Prepr. arXiv2307.01932, 2023, doi: 10.5281/zenodo.8111870.
[21] V. Da Poian et al., “Exploratory data analysis (EDA) machine learning approaches for ocean world analog mass spectrometry,” Front. Astron. Sp. Sci., vol. 10, p. 1134141, 2023, doi: 10.3389/fspas.2023.1134141.
[22] N. Ekbote, P. Dhanshetti, and S. Sakhrekar, “TECHNIQUES OF EXPLORATORY DATA ANALYSIS,” Madhya Pradesh J. Soc. Sci., vol. 28, no. 2, 2023.
[23] V. Teodorescu and L. Obreja Brașoveanu, “Assessing the Validity of k-Fold Cross-Validation for Model Selection: Evidence from Bankruptcy Prediction Using Random Forest and XGBoost,” Computation, vol. 13, no. 5, p. 127, 2025, doi: 10.3390/computation13050127.
[24] S. Ünalan, O. Günay, I. Akkurt, K. Gunoglu, and H. O. Tekin, “A comparative study on breast cancer classification with stratified shuffle split and K-fold cross validation via ensembled machine learning,” J. Radiat. Res. Appl. Sci., vol. 17, no. 4, p. 101080, 2024, doi: 10.1016/j.jrras.2024.101080.
[25] J. C. Obi, “A comparative study of several classification metrics and their performances on data,” World J. Adv. Eng. Technol. Sci., vol. 8, no. 1, pp. 308–314, 2023, doi: 10.30574/wjaets.2023.8.1.0054.
[26] H. Shafa, “Integration Of Machine Learning and Advanced Computing For Optimizing Retail Customer Analytics,” Int. J. Bus. Econ. Insights, vol. 2, no. 3, pp. 1–46, 2022, doi: 10.63125/p87sv224.
[27] A. K. Kalusivalingam, A. Sharma, N. Patel, and V. Singh, “Leveraging Random Forests and Gradient Boosting for Enhanced Predictive Analytics in Operational Efficiency,” Int. J. AI ML, vol. 3, no. 9, 2022.
[28] N. Rezki and M. Mansouri, “Machine learning for proactive supply chain risk management: Predicting delays and enhancing operational efficiency,” Manag. Syst. Prod. Eng., vol. 32, no. 3, pp. 345–356, 2024, doi: 10.2478/mspe-2024-0033.
[29] A. Rokoss, M. Syberg, L. Tomidei, C. Hülsing, J. Deuse, and M. Schmidt, “Case study on delivery time determination using a machine learning approach in small batch production companies,” J. Intell. Manuf., vol. 35, no. 8, pp. 3937–3958, 2024, doi: 10.1007/s10845-023-02290-2.
[30] M. U. Sattar, V. Dattana, R. Hasan, S. Mahmood, H. W. Khan, and S. Hussain, “Enhancing Supply Chain Management: A Comparative Study of Machine Learning Techniques with Cost--Accuracy and ESG-Based Evaluation for Forecasting and Risk Mitigation,” Sustainability, vol. 17, no. 13, p. 5772, 2025, doi: 10.3390/su17135772.
[31] H.Inaç, Y. E. Ayözen, A. Atalan, and C. Ç. Dönmez, “Estimation of postal service delivery time and energy cost with e-scooter by machine learning algorithms,” Appl. Sci., vol. 12, no. 23, p. 12266, 2022, doi: 10.3390/app122312266.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Syafrial Fachri Pane, Bargana Kukuh Raditya
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
-a.jpg)
-b.jpg)
