Gaussian Process Regression for Transportation System Estimation and Prediction Problems: The Deformation and a Hat Kernel

Article Transportation

A multi-stage stochastic optimization approach to the stop-skipping and bus lane reservation schemes

Di Huang et al.

Summary: This paper proposes a hybrid bus operational scheme combining the bus lane reservation strategy and the stop-skipping control to address the challenges of bus operation in mixed traffic environment.

TRANSPORTMETRICA A-TRANSPORT SCIENCE (2021)

Add to Collection

Article Economics

An s-shaped three-parameter (S3) traffic stream model with consistent car following relationship

Qixiu Cheng et al.

Summary: A new S3 traffic flow model was proposed in this study to represent the relationship between flow, speed, and density in highway traffic, which was calibrated using real-world traffic data. This study also derived a consistent car-following model based on the proposed S-shaped speed-density function.

TRANSPORTATION RESEARCH PART B-METHODOLOGICAL (2021)

Add to Collection

Article Economics

Estimation of urban network capacity with second-best constraints for multimodal transport systems

Zhiyuan Liu et al.

Summary: In this paper, the network capacity with second-best constraints (NCSC) is investigated for urban transport planning. A bi-level model is formulated to analyze the NCSC problem, aiming to meet specific development requirements of urban transport networks. An exact solution method is developed and verified by numerical examples, demonstrating that NCSC can be an efficient tool for transport planning and management.

TRANSPORTATION RESEARCH PART B-METHODOLOGICAL (2021)

Add to Collection

Article Economics

Macroscopic traffic flow modeling with physics regularized Gaussian process: A new insight into machine learning applications in transportation

Yun Yuan et al.

Summary: This study introduces a physics regularized machine learning (PRML) framework to enhance traffic flow modeling accuracy, especially in cases with small or noisy training datasets. By leveraging a stochastic physics regularized Gaussian process (PRGP) model and a physics regularizer based on macroscopic traffic flow models, the new approach shows superior estimation precision and robustness to noisy training datasets, surpassing previous methods.

TRANSPORTATION RESEARCH PART B-METHODOLOGICAL (2021)

Add to Collection

Article Transportation Science & Technology

Active learning for multi-objective optimal road congestion pricing considering negative land use effect

Shaopeng Zhong et al.

Summary: This study proposed a multi-objective road congestion pricing model to optimize regional accessibility, land use diversity, and flow time, and introduced an active learning optimization algorithm based on multi-objective Bayesian optimization to improve computing efficiency and address land use effect issues.

TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES (2021)

Add to Collection

Article Automation & Control Systems

Building Personalized Transportation Model for Online Taxi-Hailing Demand Prediction

Zhiyuan Liu et al.

Summary: This article proposes a personalized demand prediction model with two attention blocks to capture spatial and temporal perspectives, achieving superior prediction accuracy for large-scale online taxi-hailing demand.

IEEE TRANSACTIONS ON CYBERNETICS (2021)

Add to Collection

Article Transportation

Model on empirically calibrating stochastic traffic flow fundamental diagram

Shuaian Wang et al.

Summary: This paper addresses two shortcomings of the data-driven stochastic fundamental diagram for freeway traffic and proposes a holistic modeling framework based on the concept of mean absolute error minimization to address these issues. The established model is convex and further reformulated as a linear programming problem for efficient implementation, with experimental results validating the proposed method using real-world traffic flow data.

COMMUNICATIONS IN TRANSPORTATION RESEARCH (2021)

Add to Collection

Article Transportation

DeepTSP: Deep traffic state prediction model based on large-scale empirical data

Yang Liu et al.

Summary: This study focuses on building an effective solution for spatiotemporal data to predict the traffic state of large-scale traffic systems. The research analyzes the propagation of traffic states along the road network and proposes a deep learning architecture called Deep Traffic State Prediction (DeepTSP) to address the challenges in traffic state prediction, which is proven to effectively predict large-scale traffic states.

COMMUNICATIONS IN TRANSPORTATION RESEARCH (2021)

Add to Collection

Article Economics