Integrated dynamic analysis and predictive modeling of sustainable vehicular transport using a hybrid system dynamics-machine learning method

View at publisher Summarise with VIRAT
Authors: Arpan Paul, Siba Sankar Mahapatra
Cited by: 0
References: 45

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The dynamic behavior of factors such as product demand, travel distance, fuel price, and vehicle characteristics influencing the transportation system makes it challenging to develop a sustainable transportation model for predicting the system’s adaptability in the future. Hence, the present study proposes a novel quantitative model simulated using the system dynamics approach to visualize the influence of the considered dynamic factors of the sustainability objectives. Ensemble learning is employed across four machine learning algorithms to enhance the accuracy and reliability of predictions for the dynamic behavior of the model, considering a specific case scenario. Sobol sensitivity analysis is performed to visualize the relationships between input and output variables. Five quantitative scenario analyses have been conducted to explore the system’s future state under uncertainty. The results indicate that fuel price, vehicle speed, and product demand are crucial input parameters, while transportation costs exhibit substantial fluctuations as an output in the sustainable transportation problem.

Fields of Study & Subjects

Complex Systems and Decision Making Energy Environment and Transportation Policies Transportation Planning and Optimization

Publication details

Journal / source
Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering
Publication date
30 March 2026
Language
EN