Journal of Engineering, Project, and Production Management, 2026, 16(5), 2025-359
A Hybrid K-Means++ and Gaussian Mixture Model for High-Accuracy Travel Behavior Prediction Using ETC Data
Lecturer, School of Artificial Intelligence, Zibo Polytechnic University, Zibo, 255000, China, E-mail: DandanWangzz@outlook.com
Project Management
Received December 25, 2025; revised February 10, 2026; accepted July 7, 2026
Available online July 20, 2026
Abstract: With the continuous expansion of highway networks and the in-depth development of intelligent transportation systems, the Electronic Toll Collection (ETC) covers a large amount of user behavior data information. To enhance the accuracy of user travel prediction, this study proposes a hybrid clustering framework that employs the K-means++ algorithm to conduct efficient preliminary clustering on user travel behavior data. After determining the initial parameter range, the Gaussian Mixture Model (GMM) is used for refined iterative optimization. The accuracy in user travel prediction reached 94.83%, the recall was 93.87%, the F1 score was 94.25%, and the prediction latency was only 45.53ms, supporting 950 concurrent requests per second. In addition, on the clustering index, the silhouette coefficient of the hybrid model reached 0.64, the Calinski-Harabasz Index (CHI) was 5,861, and the Davies-Bouldin Index (DBI) was 0.68. In summary, the prediction model effectively improves the accuracy of user travel predictions and the efficiency of practical applications, and provides support and a reference for transportation network planning and personalized travel recommendations.
Keywords: Electronic toll collection (ETC); k-means++; gaussian mixture model (GMM) hybrid clustering; big data mining; traffic behavior pattern.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Wang, D. (2026). A Hybrid K-Means++ and Gaussian Mixture Model for High-Accuracy Travel Behavior Prediction Using ETC Data. Journal of Engineering, Project, and Production Management, 16(5), 2025-359.
