TY - JOUR
T1 - Digital Twin-Empowered Green Mobility Management in Next-Gen Transportation Networks
AU - Duran, Kubra
AU - Cakir, Lal Verda
AU - Fonzone, Achille
AU - Duong, Trung Q.
AU - Canberk, Berk
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2024
Y1 - 2024
N2 - Evolving transportation networks need seamless integration and effective infrastructure utilisation to form the next-generation transportation networks. Also, they should be capable of capturing the traffic flow data at the right time and promptly applying sustainable actions toward emission reduction. However, traditional transportation networks cannot handle right-time updates and act upon the requirements in dynamic conditions. Here, Digital Twin (DT) enables the development of enhanced transportation management via robust modelling and intelligence capabilities. Therefore, we propose a DT-empowered Eco-Regulation (DTER) framework with a novel twinning approach. We define a transport-specific twin sampling rate to catch right-time data in a transportation network. Besides, we perform emission prediction using Multi-Layer Perceptron (MLP), Bidirectional Long Short-Term Memory (Bi-LSTM), and BANE embeddings. We perform Laplacian matrix analysis to cluster the risk zones regarding the emissions. Thereafter, we recommend actions by setting the number of vehicle limits of junctions for high-emission areas according to the outputs of Q-learning. In summary, DTER takes control of the emission with its transport-specific twin sampling rate and automated management of transportation actions by considering the emission predictions. We note DTER achieves 19% more successful right-time data capturing, with 30% reduced query time. Moreover, our hybrid implementation of intelligent algorithms for emission prediction resulted in higher accuracy when compared to baselines. Lastly, the autonomous recommendations of DTER achieved ∼ 20% decrease in emissions by presenting an effective carbon tracing framework.
AB - Evolving transportation networks need seamless integration and effective infrastructure utilisation to form the next-generation transportation networks. Also, they should be capable of capturing the traffic flow data at the right time and promptly applying sustainable actions toward emission reduction. However, traditional transportation networks cannot handle right-time updates and act upon the requirements in dynamic conditions. Here, Digital Twin (DT) enables the development of enhanced transportation management via robust modelling and intelligence capabilities. Therefore, we propose a DT-empowered Eco-Regulation (DTER) framework with a novel twinning approach. We define a transport-specific twin sampling rate to catch right-time data in a transportation network. Besides, we perform emission prediction using Multi-Layer Perceptron (MLP), Bidirectional Long Short-Term Memory (Bi-LSTM), and BANE embeddings. We perform Laplacian matrix analysis to cluster the risk zones regarding the emissions. Thereafter, we recommend actions by setting the number of vehicle limits of junctions for high-emission areas according to the outputs of Q-learning. In summary, DTER takes control of the emission with its transport-specific twin sampling rate and automated management of transportation actions by considering the emission predictions. We note DTER achieves 19% more successful right-time data capturing, with 30% reduced query time. Moreover, our hybrid implementation of intelligent algorithms for emission prediction resulted in higher accuracy when compared to baselines. Lastly, the autonomous recommendations of DTER achieved ∼ 20% decrease in emissions by presenting an effective carbon tracing framework.
KW - Autonomous traffic management
KW - digital twin
KW - reinforcement learning
KW - twin sampling rate
UR - http://www.scopus.com/inward/record.url?scp=85207408523&partnerID=8YFLogxK
U2 - 10.1109/OJVT.2024.3484956
DO - 10.1109/OJVT.2024.3484956
M3 - Article
AN - SCOPUS:85207408523
SN - 2644-1330
VL - 5
SP - 1650
EP - 1662
JO - IEEE Open Journal of Vehicular Technology
JF - IEEE Open Journal of Vehicular Technology
ER -