Lateral shifting behavior of vehicles at horizontal curves and its influencing factors: application of LightGBM and SHAP

Samrat Debbarma; Wahengbam Ratankumar; Subhadip Biswas

doi:10.55329/ibsd7781

Authors

Samrat Debbarma National Institute of Technology Silchar, India https://orcid.org/0009-0003-7848-5896
Wahengbam Ratankumar National Institute of Technology Silchar, India https://orcid.org/0009-0002-9981-9923
Subhadip Biswas National Institute of Technology Silchar, India https://orcid.org/0000-0002-5102-0795

DOI:

https://doi.org/10.55329/ibsd7781

Keywords:

driver behavior, horizontal curve, lateral position, LightGBM, SHAP, traffic safety

Abstract

Horizontal curves are disproportionately associated with severe crashes due to increased vehicle instability and lane departure risks. This issue is particularly critical in low- and middle-income countries (LMICs), where poor lane discipline, mixed traffic, and geometric inconsistencies amplify crash potential. Despite its significance, lateral shifting (LS) behavior on horizontal curves remains understudied in LMIC contexts. This study addresses the research question: What are the key factors influencing the lateral shifting behavior of vehicles on rural horizontal curves, and how can they be modeled and interpreted effectively in LMIC conditions? Using trajectory data from 8,748 vehicles across 18 curve segments in India, an explainable machine learning framework is developed. Light Gradient Boosting Machine (LightGBM) was selected for its superior performance in classification metrics compared to other ML models. SHapley Additive exPlanations (SHAP) were integrated to interpret model outputs and quantify feature contributions, while Shannon entropy was applied to assess prediction uncertainty. Findings reveal that lane type, vehicle speed, curve radius, lateral clearance, superelevation, and traffic interactions, especially oncoming vehicles, significantly influence lateral shifting behavior. SHAP analysis uncovers nonlinear effects and interaction patterns, including a threshold response to speed and clearance. Notably, the influence of preceding vehicles differs from oncoming traffic, suggesting asymmetric behavioral responses rarely captured in prior studies. This research fills four major gaps in existing literature related to context, terrain, feature scope, and methodology. It provides data-driven insights to support safer curve design and lane departure countermeasures tailored to LMIC road environments.

Downloads

Download data is not yet available.

Author Biographies

Samrat Debbarma, National Institute of Technology Silchar, India

Samrat Debbarma is a Research Scholar at the National Institute of Technology Silchar, where he is currently pursuing a PhD in Civil Engineering with a specialization in Transportation Engineering. His research primarily focuses on traffic safety at horizontal curves, particularly in the context of hilly highways. His work integrates advanced methodologies in traffic data analysis and artificial intelligence, utilizing machine learning and statistical modeling to examine driver behavior at horizontal curves. Through this data-driven approach, he aims to contribute to the development of evidence-based strategies for enhancing traffic safety management on complex roadway geometries.

CRediT contribution: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing—original draft, Writing—review & editing.

Wahengbam Ratankumar, National Institute of Technology Silchar, India

Wahengbam Ratankumar is a M.Tech. scholar in the Department of Civil Engineering, with a specialization in Transportation Engineering, at the National Institute of Technology Silchar. His current research focuses on the analysis of vehicle lateral position at horizontal curves using naturalistic driving data and vehicle trajectory pattern. The primary objective of his work is to investigate and understand complex driver behavior under mixed-traffic conditions on horizontal curves.

CRediT contribution: Data curation, Software, Writing—original draft.

Subhadip Biswas, National Institute of Technology Silchar, India

Subhadip Biswas is an Assistant Professor in the Department of Civil Engineering at the National Institute of Technology Silchar. He received his Ph.D. degree from the Indian Institute of Technology Roorkee. Dr. Biswas’s research interests include traffic safety, traffic flow theory, and transportation systems analysis. His work involves applying advanced modeling techniques, including machine learning and statistical methods, to study driver behavior, roadway design, and crash risk assessment.

CRediT contribution: Conceptualization, Formal analysis, Methodology, Supervision, Validation, Writing—review & editing.

References

Alexei, R. T., Machemehl, R. B., & Warrenchuk, N. M. (2005). Safety impact of edge lines on rural two-lane highways,

Awasthi, D., Parti, R., & Mahajan, K. (2024). Effect of spatial relationship between curves on crash severity at horizontal curves in a mountainous terrain. Accident Analysis and Prevention, 206 DOI: https://doi.org/10.1016/j.aap.2024.107714

Ayoub, J., Yang, X. J., & Zhou, F. (2021). Modeling dispositional and initial learned trust in automated vehicles with predictability and explainability. Transportation Research Part F: Traffic Psychology and Behaviour, 77 102-116. DOI: https://doi.org/10.1016/j.trf.2020.12.015

Bassani, M., Hazoor, A., & Catani, L. (2019). What’s around the curve? A driving simulation experiment on compensatory strategies for safe driving along horizontal curves with sight limitations. Transportation Research Part F: Traffic Psychology and Behaviour, 66 273-291. DOI: https://doi.org/10.1016/j.trf.2019.09.011

Bella, F. (2013). Driver perception of roadside configurations on two-lane rural roads: Effects on speed and lateral placement. Accident Analysis and Prevention, 50 251-262. DOI: https://doi.org/10.1016/j.aap.2012.04.015

Ben-Bassat, T., & Shinar, D. (2011). Effect of shoulder width, guardrail and roadway geometry on driver perception and behavior. Accident Analysis and Prevention, 43(6), 2142-2152. DOI: https://doi.org/10.1016/j.aap.2011.06.004

Bhavna, & Biswas, S. (2022). An ANN-based framework for estimating inconsistency in lateral placement of heterogeneous traffic. Physica A: Statistical Mechanics and Its Applications, 592 DOI: https://doi.org/10.1016/j.physa.2021.126847

Cafiso, S., D’Agostino, C., & Kiec, M. (2019). Investigating safety performance of the SAFESTAR system for route-based curve treatment. Reliability Engineering and System Safety, 188 125-132. DOI: https://doi.org/10.1016/j.ress.2019.03.028

Cafiso, S., D’Agostino, C., & Persaud, B. (2018). Investigating the influence of segmentation in estimating safety performance functions for roadway sections. Journal of Traffic and Transportation Engineering (English Edition), 5(2), 129-136. DOI: https://doi.org/10.1016/j.jtte.2017.10.001

Chakraborty, M., & Gates, T. J. (2023). Relationship between horizontal curve geometry and single-vehicle crash occurrence on rural secondary highways. Transportation Research Record, DOI: https://doi.org/10.1177/03611981231208901

Charlton, S. G. (2007). The role of attention in horizontal curves: A comparison of advance warning, delineation, and road marking treatments. Accident Analysis and Prevention, 39(5), 873-885. DOI: https://doi.org/10.1016/j.aap.2006.12.007

Chen, Z., Johnsson, C., & D’Agostino, C. (2025). The effect of data transformation on the severe event prediction in road traffic using extreme value theory. Accident Analysis and Prevention, 221 DOI: https://doi.org/10.1016/j.aap.2025.108186

Das, V. R., Jayashree, M., & Rahul, S. (2016). Lateral Placement of Vehicles on Horizontal Curves. Transportation Research Procedia, 17 43-51. DOI: https://doi.org/10.1016/j.trpro.2016.11.059

Debbarma, S., & Biswas, S. (2024). Mutual Area in Lateral Distribution: Indicator of Head-On Crash Potential at Horizontal Curves on Hilly Highways DOI: https://doi.org/10.1061/JTEPBS.TEENG-8088

Debbarma, S., Ratankumar, W., & Biswas, S. (2025). Lateral shifting behavior of vehicles at sharp horizontal curves on hilly highways: insights from binomial logistic regression, Advances in Transportation Studies. 67 295-314.

Ding, H., Wang, R., Chen, T., Sze, N. N., Chung, H., & Dong, N. (2024). A hybrid approach for modeling bicycle crash frequencies: Integrating random forest based SHAP model with random parameter negative binomial regression model. Accident Analysis and Prevention, 208 DOI: https://doi.org/10.1016/j.aap.2024.107778

European Commission, (2021). Road safety thematic report – Speeding. European Road Safety Observatory,

Fitzsimmons, E. J., Kvam, V., Souleyrette, R. R., Nambisan, S. S., & Bonett, D. G. (2013). Determining vehicle operating speed and lateral position along horizontal curves using linear mixed-effects models. Traffic Injury Prevention, 14(3), 309-321. DOI: https://doi.org/10.1080/15389588.2012.701356

Glennon, J. C., & Weaver, G. D. (1971). The relationship of vehicle paths to highway curve design (No. 134-5).

Guo, M., Zhao, X., Yao, Y., Bi, C., & Su, Y. (2022). Application of risky driving behavior in crash detection and analysis. Physica A: Statistical Mechanics and Its Applications, 591 DOI: https://doi.org/10.1016/j.physa.2021.126808

Hallmark, S. L. (2012). Relationship between Speed and Lateral Position on Curves, www.intrans.iastate.edu/

Hancock, J. T., & Khoshgoftaar, T. M. (2020). CatBoost for big data: an interdisciplinary review. Journal of Big Data, 7(1), DOI: https://doi.org/10.1186/s40537-020-00369-8

Havránek, P., Zůvala, R., Špaňhel, J., Herout, A., Valentová, V., & Ambros, J. (2020). How does road marking in horizontal curves influence driving behaviour?. European Transport Research Review, 12(1), DOI: https://doi.org/10.1186/s12544-020-00425-7

Hicks, S. A., Strümke, I., Thambawita, V., Hammou, M., Riegler, M. A., Halvorsen, P., & Parasa, S. (2022). On evaluation metrics for medical applications of artificial intelligence. Scientific Reports, 12(1), DOI: https://doi.org/10.1038/s41598-022-09954-8

Jalal, M., Kamal, M., & Zafar, A. (2024). ChemCarcinoPred: Carcinogenicity Prediction of Small Drug-Like Molecules Using LightGBM and Molecular Fingerprints. Biophysical Reviews and Letters, 19(04), 409-424. DOI: https://doi.org/10.1142/S1793048023410035

Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). LightGBM: A Highly Efficient Gradient Boosting Decision Tree. 31st Conference on Neural Information Processing Systems.

Krammes, R. A., & Tyer, K. D. (1991). Post Mounted Delineators and RPMs: Their Effect on Vehicle Operations at Horizontal Curves on Two-Lane Rural Highways.

Li, J. (2024). Area under the ROC Curve has the most consistent evaluation for binary classification. PLoS ONE, 19(12), DOI: https://doi.org/10.1371/journal.pone.0316019

Lundberg, S. M., Erion, G., Chen, H., DeGrave, A., Prutkin, J. M., Nair, B., Katz, R., Himmelfarb, J., Bansal, N., & Lee, S. I. (2020). From local explanations to global understanding with explainable AI for trees. Nature Machine Intelligence, 2(1), 56-67. DOI: https://doi.org/10.1038/s42256-019-0138-9

Lundberg, S. M., & Lee, S.-I. (2017). A Unified Approach to Interpreting Model Predictions. 31st Conference on Neural Information Processing Systems,

Maljković, B., & Cvitanić, D. (2016). Evaluation of design consistency on horizontal curves for two-lane state roads in terms of vehicle path radius and speed. Baltic Journal of Road and Bridge Engineering, 11(2), 127-135. DOI: https://doi.org/10.3846/bjrbe.2016.15

Mauriello, F., Montella, A., Pernetti, M., & Galante, F. (2018). An exploratory analysis of curve trajectories on two-lane rural highways. Sustainability (Switzerland), 10(11), DOI: https://doi.org/10.3390/su10114248

MoRTH, (2022). Road accidents in India.

Mosca, E., Szigeti, F., Tragianni, S., Gallagher, D., & Groh, G. (2022). SHAP-Based Explanation Methods: A Review for NLP Interpretability. Proceedings of the 29th International Conference on Computational Linguistics, 4593-4603.

NHTSA, (2016). Traffic safety facts.

Owusu-Adjei, M., Ben Hayfron-Acquah, J., Frimpong, T., & Abdul-Salaam, G. (2023). Imbalanced class distribution and performance evaluation metrics: A systematic review of prediction accuracy for determining model performance in healthcare systems. PLOS Digital Health, 2(11), DOI: https://doi.org/10.1371/journal.pdig.0000290

Ponsam, J. G., Bella Gracia, S. V. J., Geetha, G., Karpaselvi, S., & Nimala, K. (2021). Credit Risk Analysis using LightGBM and a comparative study of popular algorithms. Proceedings of the 2021 4th International Conference on Computing and Communications Technologies (ICCCT 2021), 634-641. DOI: https://doi.org/10.1109/ICCCT53315.2021.9711896

Saini, H. K., & Biswas, S. (2021). Estimating lateral placement and lane indiscipline of urban mixed traffic of a developing country: An ann-assisted approach. Canadian Journal of Civil Engineering, 48(11), 1571-1581. DOI: https://doi.org/10.1139/cjce-2020-0250

Schneider, W. H., Savolainen, P. T., & Zimmerman, K. (2009). Driver injury severity resulting from single-vehicle crashes along horizontal curves on rural two-lane highways. Transportation Research Record, 2102 85-92. DOI: https://doi.org/10.3141/2102-11

Shannon, C. E. (1948). A Mathematical Theory of Communication. The Bell System Technical Journal, 27 623-656. DOI: https://doi.org/10.1002/j.1538-7305.1948.tb00917.x

Shapley, L. S. (1953). A value for n-person games. Contributions to the Theory of Games: Vol. II, 307-317. DOI: https://doi.org/10.1515/9781400881970-018

Sharma, S., Gautam, A., Debbarma, S., Biswas, S., & Sharma, S. (2025). Lateral placement distribution of vehicles at horizontal curves on two-way undivided roads, International Journal of Crashworthiness, 0(0), 1-15. DOI: https://doi.org/10.1080/13588265.2025.2492990

Spacek, P. (2005). Track behavior in curve areas: Attempt at typology. Journal of Transportation Engineering, 131(9), 669-676. DOI: https://doi.org/10.1061/(ASCE)0733-947X(2005)131:9(669)

Stodart, B. P., Donnell, E. T., & Asce, M. (2008). Speed and Lateral Vehicle Position Models from Controlled Nighttime Driving Experiment. DOI: https://doi.org/10.1061/(ASCE)0733-947X(2008)134:11(439)

Tiwari, G., Fazio, J., & Pavitavas, S. (2000). Passenger Car Units for Heterogeneous Traffic Using a Modified Density Method, 246-257.

Trivedi, M. M., & Gor, R. (2017). Influence of lane discipline on traffic flow characteristics at Maninagar level crossing. International Journal of Novel Research and Development, 2(5), www.ijnrd.org

Wen, X., Xie, Y., Wu, L., & Jiang, L. (2021). Quantifying and comparing the effects of key risk factors on various types of roadway segment crashes with LightGBM and SHAP. Accident Analysis and Prevention, 159 DOI: https://doi.org/10.1016/j.aap.2021.106261

WHO, (2023). Global status report on road safety,

Wu, J., & Xu, H. (2017). Driver behavior analysis for right-turn drivers at signalized intersections using SHRP 2 naturalistic driving study data. Journal of Safety Research, 63 177-185. DOI: https://doi.org/10.1016/j.jsr.2017.10.010

Xin, C., Wang, Z., Lee, C., & Lin, P. S. (2017). Modeling safety effects of horizontal curve design on injury severity of single-motorcycle crashes with mixed-effects logistic model. Transportation Research Record, 2637(1), 38-46. DOI: https://doi.org/10.3141/2637-05

Xue, H., Guo, P., Li, Y., & Ma, J. (2024). Integrating visual factors in crash rate analysis at intersections: An AutoML and SHAP approach towards cycling safety. Accident Analysis and Prevention, 200 DOI: https://doi.org/10.1016/j.aap.2024.107544

Yastremska-Kravchenko, O., Laureshyn, A., D’Agostino, C., & Varhelyi, A. (2022). What constitutes traffic event severity in terms of human danger perception?. Transportation Research Part F: Traffic Psychology and Behaviour, 90 22-34. DOI: https://doi.org/10.1016/j.trf.2022.08.001

Zhang, Y., Chen, Y., Gu, X., Sze, N. N., & Huang, J. (2023). A proactive crash risk prediction framework for lane-changing behavior incorporating individual driving styles. Accident Analysis and Prevention, 188 DOI: https://doi.org/10.1016/j.aap.2023.107072