Transportation Mode Choice for Shopping Trip Using Random Forest Algorithm: Pemilihan Moda Transportasi Untuk Perjalanan Belanja  Menggunakan Algoritma Random Forest

Kezia Najoan; Semuel Rompis; Lucia Lefrandt

doi:10.35793/jtek.v14i1.62304

Authors

Kezia Najoan Universitas Sam Ratulangi
Semuel Rompis
Lucia Lefrandt

DOI:

https://doi.org/10.35793/jtek.v14i1.62304

Keywords:

Artificial Intelligence , Mode Choice, Online Transportation, Private Vehicle, Random Forest

Abstract

Abstract — Manado City, home to over 460,000 residents and anchored by the Megamas business and shopping district, faces significant mobility challenges and complex traffic congestion. The rise of online transport services has provided residents with new options, yet private vehicles remain the dominant choice due to comfort and scheduling flexibility. This study aims to describe travel‐actor characteristics in Manado, identify key factors influencing mode choice, and apply a Random Forest Classifier to predict preferences between private vehicles and online transport services. Primary data were collected from 203 respondents via online questionnaires and field interviews, covering demographics, trip characteristics, and choice drivers. Data were cleaned, encoded, and split into 80% training and 20% testing subsets. A default Random Forest model (100 trees) showed overfitting (100% train vs. 78% test). Hyperparameter tuning (GridSearchCV, 5-fold) improved test accuracy to 80.5%. Removing the least important feature (“trip distance”) further boosted test accuracy to 85.4% with average F₁-score >85%. Feature importance ranked vehicle ownership (25.4%) and primary reason (22.3%) as top predictors. Recommendations include expanding fleet coverage in suburban areas, adjusting pricing for lower-income groups, and intensifying public education on online transport benefits to alleviate congestion. Future research could explore multimodal integrations and real-time data analytics to optimize urban mobility systems.

Key words— Artificial Intelligence, Mode Choice, Online Transportation, Private Vehicle, Random Forest

Abstrak — Kota Manado, dengan lebih dari 460.000 penduduk dan kawasan Megamas sebagai pusat bisnis dan perbelanjaan, menghadapi tantangan mobilitas tinggi dan kemacetan kompleks. Munculnya layanan transportasi online menawarkan alternatif, namun kendaraan pribadi masih mendominasi pilihan karena kenyamanan dan fleksibilitas waktu. Penelitian ini bertujuan menggambarkan karakteristik pelaku perjalanan di Manado, mengidentifikasi faktor penentu pilihan moda, dan menerapkan Random Forest Classifier untuk memprediksi preferensi antara kendaraan pribadi dan transportasi online. Data primer diperoleh dari 203 responden melalui kuesioner daring dan wawancara lapangan, mencakup demografi, karakteristik perjalanan, dan pendorong pilihan. Data dibersihkan, dikodekan, dan dibagi 80% training–20% testing. Model Random Forest default (100 pohon) menunjukkan overfitting (akurasi train 100% vs. test 78%). Hyperparameter tuning (GridSearchCV 5-fold) meningkatkan akurasi test menjadi 80,5%. Menghapus fitur terendah (“jarak perjalanan”) menaikkan akurasi test menjadi 85,4% dengan F₁-score rata-rata >85%. Feature importance menempatkan kepemilikan kendaraan (25,4%) dan alasan utama (22,3%) sebagai prediktor utama. Rekomendasi meliputi perluasan jangkauan armada di zona pinggiran, penyesuaian tarif untuk kelompok berpendapatan rendah, dan edukasi publik tentang keunggulan transportasi daring untuk mengurangi kemacetan. Penelitian lanjutan dapat mengeksplorasi integrasi multimoda dan analitik data real-time untuk mengoptimalkan sistem mobilitas perkotaan.

Kata kunci — Kecerdasan Buatan, Kendaraan Pribadi, Pemilihan Moda, Pohon Keputusan, Transportasi Online

References

Badan Pusat Statistik Manado, Kota Manado Dalam Angka, vol. 21. Manado: BPS Kota Manado, 2024.

S. Y. R. Rompis, “Karakteristik Pemilihan Moda di Kota Manado Dengan Metode Multinomial Logit,” Jurnal Penelitian Jalan dan Jembatan, vol. 1, no. 1, 2021, [Online]. Available: https://doi.org/10.59900/ptrkjj.v1i1.25

J. Matulende, L. Lefrandt, and S. Pandey, “Pengaruh Angkutan Online Terhadap Pemilihan Moda Transportasi Publik Di Kota Manado(Studi Kasus: Trayek Sumompo-Pusat Kota),” TEKNO, vol. 20, no. 81, 2022, [Online]. Available: https://doi.org/10.35793/jts.v20i81.43496

R. L. Hermaputi and C. Hua, “Decoding Jakarta Women’s Non-Working Travel-Mode Choice: Insights from Interpretable Machine-Learning Models,” Sustainability (Switzerland), vol. 16, no. 19, Oct. 2024, doi: 10.3390/su16198454.

L. Lefrandt and M. Kumaat, “Characteristics of Transportation Mode Selection in Manado Maritime City,” International Journal of Marine Engineering Innovation and Research, vol. 9, no. 4, pp. 2548–1479, 2024.

M. Maranatha, M. Audie, L. E. Rumayar, and L. Jefferson, “Model Pemilihan Moda Angkutan Umum dan Transportasi Online di Kota Tomohon (Studi Kasus: Pelajar di Kota Tomohon),” Jurnal Sipil Statik, vol. 8, no. 6, pp. 911–924, 2020.

N. Fahriza et al., “Travel Mode Choice Modeling: Predictive Efficacy between Machine Learning Models and Discrete Choice Model,” The Open Transportation Journal, vol. 15, 2021, doi: 10.2174/18744478021150102.

C. R. Sekhar, Minal, and E. Madhu, “Mode Choice Analysis Using Random Forrest Decision Trees,” in Transportation Research Procedia, Elsevier B.V., 2016, pp. 644–652. doi: 10.1016/j.trpro.2016.11.119.

H. Zhang, L. Zhang, Y. Liu, and L. Zhang, “Understanding Travel Mode Choice Behavior: Influencing Factors Analysis and Prediction with Machine Learning Method,” Sustainability (Switzerland), vol. 15, no. 14, Jul. 2023, doi: 10.3390/su151411414.

J. J. M. D’Cruz, A. P. Alex, and V. S. Manju, “MODE CHOICE ANALYSIS OF SCHOOL TRIPS USING RANDOM FOREST TECHNIQUE,” Archives of Transport, vol. 63, no. 3, pp. 39–48, 2022, doi: 10.5604/01.3001.0015.9175.

N. F. Mohd Ali, A. F. Mohd Sadullah, A. P. P. Abdul Majeed, M. A. Mohd Razman, and R. M. Musa, “The identification of significant features towards travel mode choice and its prediction via optimised random forest classifier: An evaluation for active commuting behavior,” J Transp Health, vol. 25, Jun. 2022, doi: 10.1016/j.jth.2022.101362.

H. Zhang and N. Liu, “Behavioral Analysis of Urban Travel Mode Selection Based on Random Forest Algorithm,” International Journal of Multiphysics, vol. 18, no. 3, 2024.

H. A. Kalantari, S. Sabouri, S. Brewer, R. Ewing, and G. Tian, “Machine Learning in Mode Choice Prediction as Part of MPOs’ Regional Travel Demand Models: Is It Time for Change?,” Sustainability (Switzerland), vol. 17, no. 8, Apr. 2025, doi: 10.3390/su17083580.

L. Breiman, “Random Forests,” 2001.

F. Pedregosa FABIANPEDREGOSA et al., “Scikit-learn: Machine Learning in Python Gaël Varoquaux Bertrand Thirion Vincent Dubourg Alexandre Passos PEDREGOSA, VAROQUAUX, GRAMFORT ET AL. Matthieu Perrot,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011, [Online]. Available: http://scikit-learn.sourceforge.net.