Article
Economics
Xiaolei Wang, Jun Wang, Lei Guo, Wei Liu, Xiaoning Zhang
Summary: A new modeling approach for ridesharing user equilibrium (RUE) was proposed, which transforms the problem into a convex programming problem by redefining feasible driver trajectories and ridesharing market equilibrium conditions. The algorithm effectively avoids path enumeration and can be implemented on large networks, with theoretical analysis and numerical demonstrations on the impact of problem size on computational efficiency.
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
(2021)
Article
Computer Science, Interdisciplinary Applications
Sam O'Neill, Ovidiu Bagdasar, Stuart Berry, Nicolae Popovici, Ramachandran Raja
Summary: This paper presents a method of considering multiple objectives simultaneously in selfish routing of network flow. By manipulating free parameters such as speed limits, the behavior of road users is coerced to reconcile conflicts between multiple objectives. The results show that small parameter adjustments can lead to solutions that Pareto dominate other solutions.
MATHEMATICS AND COMPUTERS IN SIMULATION
(2022)
Article
Energy & Fuels
Quan Yuan, Yujian Ye, Yi Tang, Yuanchang Liu, Goran Strbac
Summary: This paper proposes a novel deep learning based surrogate modeling method for effective modeling and optimization of EV flows and charging demand. Case studies demonstrate that the proposed method outperforms existing methods in both solution accuracy and computational performance. Coordinated spatial optimization of EV flows and charging demand benefits the operation of both TN and PDN.
Article
Engineering, Electrical & Electronic
Yan Wang, Qun Chen
Summary: This study analyzes the parking allocation problem in a multidestination multiple parking lot system, considering factors that affect driver's parking choice, and establishes a mathematical programming model. The proposed model is validated through a numerical example and discusses the impact of dynamic parking fees on driver's choice and parking lot utilization balance. The model is of significance to parking facility location optimization and dynamic parking pricing strategy.
IEEE INTELLIGENT TRANSPORTATION SYSTEMS MAGAZINE
(2022)
Article
Thermodynamics
Ze Zhou, Zhitao Liu, Hongye Su, Liyan Zhang
Summary: Dynamic wireless charging technology can alleviate range anxiety and promote the widespread use of electric vehicles. It is important to consider it as a significant charging method in future urban scenarios. This study proposes a planning strategy for the coexistence of static and dynamic charging facilities in order to maximize the comprehensive performance of the power-traffic system.
Article
Economics
Ruqing Huang, Lee D. Han, Zhongxiang Huang
Summary: This paper presents an unconventional equilibrium flow model to analyze travelers' route choice behavior, using path residual capacity as the quantity signal. The proposed model and solution algorithm are shown to be feasible and effective, and can capture route choice behaviors that have not been modeled in previous studies.
TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW
(2022)
Article
Economics
Terry L. Friesz, Ke Han, Amir Bagherzadeh
Summary: This paper presents sufficient conditions for convergence of projection and fixed-point algorithms used to compute dynamic user equilibrium with elastic travel demand, without the need for strongly monotone increasing path delay operators. Instead, weakly monotone increasing path delay operators and strongly monotone decreasing inverse demand functions are assumed. The Lipschitz continuity of path delay is a mild regularity condition, allowing for convergence even with nonmonotone delay operators under certain conditions.
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
(2021)
Article
Economics
Tongfei Li, Min Xu, Huijun Sun, Jie Xiong, Xueping Dou
Summary: In this study, a generalized stochastic user equilibrium model is developed to analyze travelers' mode and route choice behavior in urban traffic systems with ridesharing programs. The proposed model considers travelers' heterogeneity in terms of car ownership and value of time, and their limited perceived information based on the stochastic user equilibrium principle. The decision-making problem of ridesharing compensation is also addressed, aiming to minimize total travel cost and vehicular air pollution emissions. A bi-objective optimization model and two single-objective optimization models are proposed, and a genetic algorithm is used to generate Pareto-optimal solutions. Numerical experiments demonstrate the effectiveness of the proposed model and algorithm in mitigating traffic congestion and pollution emissions.
TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW
(2023)
Article
Transportation Science & Technology
Nam H. Hoang, Manoj Panda, Hai L. Vu, Dong Ngoduy, Hong K. Lo
Summary: This study focuses on a transport network with two types of users, selfish and cooperative. Selfish users aim to minimize their travel time, while cooperative users aim to maximize their class's aggregate throughput or minimize their total travel time. A new framework is proposed to study the route choices and network performance of the two classes of users.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2023)
Article
Transportation Science & Technology
Mohammad Hadi Mansourianfar, Ziyuan Gu, S. Travis Waller, Meead Saberi
Summary: The paper proposes a joint routing and pricing control scheme to incentivize CAVs to seek system-optimal routing, reducing total system travel time and discouraging congestion by human-driven vehicles in city centers.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2021)
Article
Computer Science, Interdisciplinary Applications
Xiang Zhang, Steven Travis Waller, Dung-Ying Lin
Summary: This study is the first in the literature to examine the Braess paradox considering parking behavior in the autonomous vehicle (AV) environment and model the network design problem for the autonomous transportation system (NDP-ATS). It shows the existence of two distinct Braess paradoxes in AV traffic networks and develops a bi-level programming model to avoid the deterioration caused by these paradoxes. The results highlight the efficacy of the modeling framework for infrastructure development and policy assessment for AV traffic.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2023)
Article
Environmental Studies
D. Kang, F. Hu, M. W. Levin
Summary: This study examines the impacts of induced AV trips on the transportation network and proposes a solution algorithm. Test results demonstrate that the use of AVs increases average travel time and allows for the repurposing of parking spaces.
TRANSPORTATION RESEARCH PART D-TRANSPORT AND ENVIRONMENT
(2022)
Article
Computer Science, Software Engineering
Alexander Y. Krylatov
Summary: Nowadays, traffic flow assignment has become an interdisciplinary topic that concerns multiple research areas and branches of science. This work focuses on the mathematical and computational aspects of the equilibrium traffic assignment problem when route flows are considered as decision variables. The developed fixed-point mapping with the explicit contraction operator proves to equilibrate journey times on feasible routes efficiently. Furthermore, the sequential path-equilibration algorithm shows higher accuracy in user-equilibrium traffic assignment solutions compared to the best ones known previously.
OPTIMIZATION METHODS & SOFTWARE
(2023)
Article
Computer Science, Interdisciplinary Applications
Xiang Zhang, Edward Robson, S. Travis Waller
Summary: This study investigates the potential positive or negative societal impacts of AVs' expected travel and parking behavior, developing an integrated transport and economic model and conducting a case study in Sydney to demonstrate the significant losses of social welfare that can result from AV parking patterns.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2021)
Article
Engineering, Multidisciplinary
Tomas Potuzak, Frantisek Kolovsky
Summary: This paper describes a technique for predicting traffic flows in individual roads of a road traffic network and presents the parallelization process of the technique, along with the convergence, usability, and speed tests conducted on real road traffic networks.
AIN SHAMS ENGINEERING JOURNAL
(2022)
Article
Engineering, Mechanical
Yixuan Liu, Chen Jiang, Xiaoge Zhang, Zissimos P. Mourelatos, Dakota Barthlow, David Gorsich, Amandeep Singh
Summary: This article introduces a bio-inspired approach for multivehicle mission planning of off-road autonomous ground vehicles in dynamic environments. It analyzes vehicle reliability using physics-based simulations and identifies optimal paths using a combination of Physarum algorithm and navigation mesh.
JOURNAL OF MECHANICAL DESIGN
(2022)
Article
Computer Science, Artificial Intelligence
Xiaoge Zhang, Sankaran Mahadevan, Felix T. S. Chan
Summary: This paper presents a framework based on uncertainty quantification to enhance the interpretability of deep learning models. By propagating uncertainty to model predictions and optimizing pixels using entropy and differential evolution, the model interpretability is effectively improved.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Engineering, Industrial
Xiaoge Zhang, Prabhakar Srinivasan, Sankaran Mahadevan
Summary: This paper applies data-mining and deep learning techniques to analyze NTSB accident investigation reports, developing classification models to predict adverse events. The deep learning models help safety professionals audit, review, and analyze accident reports, performing scenario analyses.
Article
Computer Science, Artificial Intelligence
Xiaoge Zhang, Felix T. S. Chan, Chao Yan, Indranil Bose
Summary: This paper provides a systematic and comprehensive overview of the various risks that may arise in AI/ML systems, and emphasizes the need for research on developing a risk management framework.
DECISION SUPPORT SYSTEMS
(2022)
Article
Computer Science, Hardware & Architecture
Xiaoge Zhang, Zhen Hu, Sankaran Mahadevan
Summary: This article investigates a resilience-based design optimization approach for configuring logistics service centers. The approach considers the impact of potential disruptive events and aims to enhance the system's ability to withstand such events through optimized decision variables. An adaptive importance sampling method is used to tackle the complex optimization problem, and the effectiveness of the proposed approach is demonstrated through a numerical example.
IEEE TRANSACTIONS ON RELIABILITY
(2022)
Article
Computer Science, Artificial Intelligence
Cho Yin Yiu, Kam K. H. Ng, Xinyu Li, Xiaoge Zhang, Qinbiao Li, Hok Sam Lam, Man Ho Chong
Summary: Teams composed of aviation professionals are crucial for maintaining a safe and efficient airport environment. This research evaluates the impact of an enhanced communication protocol on cognitive workload under adverse weather conditions and develops a human-centered classification model for identifying hazardous meteorological conditions. The findings indicate that reduced visibility significantly increases subjective workload, but inclusion of turning direction information in communications does not intensify cognitive workload. The proposed Bayesian neural network-based classification model outperforms other algorithms in identifying potentially hazardous weather conditions.
ADVANCED ENGINEERING INFORMATICS
(2022)
Article
Engineering, Civil
Yingxiao Kong, Xiaoge Zhang, Sankaran Mahadevan
Summary: This paper focuses on the hard landing problem, which is one of the riskiest phases of a flight. A probabilistic predictive model is built using machine learning and Bayesian neural network approach to forecast the aircraft's vertical speed at touchdown. The model is validated using test flights and shows satisfactory performance.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Transportation Science & Technology
Xiaoge Zhang, Sanqiang Zhong, Sankaran Mahadevanb
Summary: This paper introduces a machine learning model for predicting the trajectories of ground objects on the airport surface, aiming to reduce collision events. The model utilizes a spatial-temporal graph convolutional neural network (STG-CNN) and other techniques to forecast the movement of objects and defines a separation distance-based metric for assessing safety. The model's performance is validated at two airports, demonstrating its superiority over an alternative approach.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2022)
Article
Engineering, Industrial
Taotao Zhou, Xiaoge Zhang, Enrique Lopez Droguett, Ali Mosleh
Summary: In this paper, a PINN-based framework is proposed to assess the reliability of multi-state systems. The framework uses machine learning to convert the reliability assessment into a problem and solves the issue of gradient imbalance and establishes a continuous latent function. Experimental results show that the framework performs well in MSS reliability assessment.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Correction
Computer Science, Interdisciplinary Applications
Adam Thelen, Xiaoge Zhang, Olga Fink, Yan Lu, Sayan Ghosh, Byeng D. Youn, Michael D. Todd, Sankaran Mahadevan, Chao Hu, Zhen Hu
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2023)
Review
Computer Science, Interdisciplinary Applications
Adam Thelen, Xiaoge Zhang, Olga Fink, Yan Lu, Sayan Ghosh, Byeng D. Youn, Michael D. Todd, Sankaran Mahadevan, Chao Hu, Zhen Hu
Summary: Digital twin, an emerging technology in the era of Industry 4.0, is comprehensively modeling the physical world as interconnected digital models. This paper provides a literature review of digital twin trends, analyzes digital twin modeling and enabling technologies, and offers perspectives on the future trajectory of digital twin technology.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2022)
Review
Computer Science, Interdisciplinary Applications
Adam Thelen, Xiaoge Zhang, Olga Fink, Yan Lu, Sayan Ghosh, Byeng D. Youn, Michael D. Todd, Sankaran Mahadevan, Chao Hu, Zhen Hu
Summary: Digital twin, as an emerging technology in the industry 4.0 era, is drawing unprecedented attention due to its potential in optimizing various processes. In this second part of the paper, the focus is on reviewing the key enabling technologies of digital twins, including uncertainty quantification, optimization methods, open-source datasets and tools. A case study of a battery digital twin is presented to illustrate the modeling and twinning methods discussed in the review. The code and preprocessed data for generating the case study results are available on Github.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2023)
Article
Engineering, Manufacturing
Jingchang Li, Xiaoge Zhang, Qi Zhou, Felix T. S. Chan, Zhen Hu
Summary: This paper proposes a feature-level multi-sensor fusion approach that combines acoustic emission signals with photodiode signals to achieve quality monitoring in selective laser melting (SLM) technology. By developing an off-axial monitoring system to capture process signatures and using a convolutional neural network to extract and fuse features from two sensors, the proposed approach outperforms several baseline models in quality monitoring.
JOURNAL OF MANUFACTURING PROCESSES
(2022)
Article
Engineering, Multidisciplinary
Yiming Zhang, Dingyang Zhang, Xiaoge Zhang, Lemiao Qiu, Felix T. S. Chan, Zili Wang, Shuyou Zhang
Summary: This paper proposes a Guided Probabilistic Reinforcement Learning (Guided-PRL) model to minimize the life-cycle cost of multi-component systems maintenance. The Guided-PRL model improves upon traditional Actor-Critic models by introducing a guided sampling scheme and Bayesian models for uncertainty quantification.
APPLIED MATHEMATICAL MODELLING
(2023)
Article
Computer Science, Hardware & Architecture
Wenting Yi, Wai Kit Chan, Hiu Hung Lee, Steven T. Boles, Xiaoge Zhang
Summary: This article introduces the importance of uncertainty quantification in mission-critical engineering applications and presents a methodology that seamlessly integrates a spectral-normalized neural Gaussian process (SNGP) module into GoogLeNet for accurately detecting defects in steel wire ropes. The methodology consists of three steps, including collecting raw magnetic flux leakage (MFL) signals, transforming the signals into 2-D images using Gramian angular field, and integrating SNGP into GoogLeNet with spectral normalization (SN) and Gaussian process (GP) layers. Comparative evaluations demonstrate the advantages of the developed methodology in classifying SWR defects and identifying out-of-distribution (OOD) SWR instances.
IEEE TRANSACTIONS ON RELIABILITY
(2023)
