Journal of Advanced Transportation最新文献

This study introduces a novel univariate probability density (UPD) model leveraging partially monotone neural networks to analyze activity durations, with a specific focus on shopping trips by noncommuters in Shanghai. The proposed method ensures the monotonicity of the cumulative distribution function (CDF) with respect to time while enabling flexible modeling of complex distributions influenced by exogenous variables. Simulation experiments validate the model’s robustness and accuracy in capturing distributional patterns and variable effects. Empirical analysis using the 2019 Shanghai Household Travel Survey data demonstrates the model’s capability to reveal nuanced relationships between shopping durations and demographic, household, and locational factors. The results provide valuable insights into activity-based modeling and inform urban planning, transportation systems, and policy-making. By enabling realistic sampling and robust scenario analysis, this approach establishes a flexible, data-driven framework for studying activity durations.

Understanding the conditions that affect takeover time (TOT) in conditional autonomous driving systems remains a challenging issue. The takeover process requires a seamless transition of control from the autonomous system to the driver when the system encounters situations it cannot manage. This study examines the effects of traffic conditions, road geometry, and weather on TOT using a linear mixed model to quantify their influence. Preliminary findings indicate that factors such as rain, gender, and age significantly extend control transition duration. These insights highlight the need for personalized designs in automated driving systems (ADSs) and takeover request protocols to accommodate diverse driver characteristics and environmental conditions. While the research utilizes a driving simulator, suggesting the need for field validation, it offers a foundational understanding that can enhance the safety and efficiency of conditional automation systems. This study contributes to safer ADS design and supports the commercial viability of conditional autonomous vehicles.

Vehicle lateral motion control is one of the critical issues in intelligent vehicle control. We design a vehicle lateral motion controller by combining the adaptive fading Sage–Husa Kalman filter (AFSH-KF) with the robust model predictive algorithm to address the problem of vehicle lateral motion control. Due to the influence of process and measurement noise on the estimation results, the AFSH-KF is employed to estimate the vehicle state parameters to improve the estimation accuracy and compared with the Kalman filter (KF). Simultaneously considering the influence of the uncertainties or perturbations appearing in the feedback loop (vehicle state parameters) on vehicle lateral motion control, a robust model predictive controller (RMPC) is designed for vehicle lateral motion. The performance of the designed controller is verified by co-simulating with MATLAB/Simulink and CarSim in double-lane, S-shape, and Fishhook conditions. The results show that the AFSH-KF can effectively estimate the states (yaw rate and sideslip angle) of the vehicle. Compared to the MPC controller, the RMPC controller significantly reduced the maximum and mean square error of the lateral deviation of the vehicle tracking target trajectory at different speeds.

In the oversaturated metro system, the mismatch between supply and demand leads to unequal allocation of train capacity at different stations, resulting in a transportation inequity issue. This paper proposes a collaborative optimization method to use train carriage flexible release strategy and passenger flow control strategy, which is described as a mixed-integer nonlinear programming (MINLP) model considering the trade-off between equity and efficiency. To solve this model, it is reformulated into a mixed-integer linear programming (MILP) model, which is solved by the GUROBI solver. An efficient variable neighborhood search algorithm is then proposed to find a high-quality solution to the proposed problem. Finally, two sets of numerical experiments, including a small-scale case and a real-world case of Chengdu metro system, are conducted to verify the proposed model. The experimental results show that the train release scheme and passenger flow control scheme generated by our proposed method can perform well on the trade-off between equity and efficiency.

Passenger pushing behavior during emergency evacuations on roll-on/roll-off (Ro-Ro) passenger ships is a critical yet overlooked factor in evacuation modeling. This study investigates the impact of pushing behavior on evacuation dynamics by employing an improved social force model (SFM) that integrates pushing forces and the ship’s inclination angle. Four evacuation scenarios are simulated to evaluate the impacts of pushing behavior and falling incidents. Results show that (1) moderate pushing can slightly shorten evacuation time without significantly increasing the risk of falling; (2) excessive pushing induces localized congestion, elevates the probability of falls, and ultimately prolongs evacuation time—under severe pushing conditions, total evacuation time increased by 45.4% compared with the no-pushing baseline; and (3) ship inclination significantly affects passenger stability, particularly near exit bottlenecks and in narrow passages. The findings enhance the realism of evacuation simulations and provide practical insights for optimizing crowd management strategies on Ro-Ro passenger ships.

This study was conducted to ensure traffic continuity at an adaptive signalized intersection by developing a SUMO-based digital twin of the Heybe Intersection in Antalya, using real traffic data obtained from the Antalya Traffic Control Center (covering 165 days of observations). To address potential sensor failure scenarios, a solution integrating traffic forecasting and reinforcement learning was developed. After applying data cleaning techniques, multiple deep learning models were trained to forecast traffic volumes, and their outputs were used to generate an origin-destination (O/D) matrix that served as input to a Deep Q-Learning (DQL) control model. Three scenarios were evaluated in the simulation: (i) baseline adaptive signal control under normal operating conditions, (ii) the existing system under sensor failure reverting to a fixed-time plan, and (iii) the proposed DQL-based intersection management. Results demonstrated that, under sensor failure conditions, the DQL-based system achieved substantial improvements compared to the fixed-time baseline: the average delay was reduced by 61.3%, the average speed increased by 134.6%, and the level of service improved from E to B. These findings highlight the potential of integrating forecasting models with DQL to enhance the resilience of smart intersections against sensor malfunctions.

The purpose of launching railway group tickets for railway enterprises is twofold: (1) increase revenue; (2) attract new users to travel by railway. In order to study how to achieve the above goals through price strategies for group tickets, this paper proposes an optimization approach for railway group ticket pricing in a scenario of multitrains. First, based on the consistent preference of passengers for group tickets, we model the decision-making process of existing passengers purchasing group tickets and calculate the required quantitative boundary of existing passengers for selling out group tickets in order of priority. Then, under the constraints of stochastic demand and shared seat quota between group tickets and individual tickets, a multiobjective nonlinear optimization model with the objectives of maximizing both total expected revenue and expected sales of new users is constructed and solved. The analysis results reveal that there is no unique optimal solution simultaneously maximizing the two objectives. Increasing expected revenue will sacrifice the goal of attracting more incremental passengers to take trains. Limited by the fixed seat allocation, a scientific moderate discount scheme on group tickets can increase the total expected revenue. At this time, selling both group tickets and individual tickets yields higher revenue than only selling individual tickets, thus verifying the rationality of the mixed sales strategy of group tickets and individual tickets. Furthermore, we find an indicator named “elasticity of existing passengers” that has a critical impact on the expected revenue. Railway enterprises should take measures to incentivize the marketing enthusiasm of third-party sales agencies to minimize the elasticity of existing passengers to achieve greater revenue.

Shared micromobility services are experiencing rapid growth, particularly in addressing last-mile transportation needs. The most crucial questions focus on identifying the determinants of user behavior and the factors driving demand for micromobility vehicles. Investigating this topic is thus essential for meeting the demand of micromobility vehicles, ensuring their dynamic and flexible deployment, and optimizing overall system planning. In this study, demand forecasting was performed using a shared electric scooter (e-scooter) dataset and by comparing 19 distinct machine learning (ML) and deep learning (DL) algorithms, including traditional ML algorithms, neural network–based (NN) models , ANN and metaheuristic hybrid models, and ensemble models. Algorithm performance, evaluated using R² and RMSE metrics, shows that boosting and hybrid models significantly outperform traditional algorithms. In this study, the algorithms were compared not only with RMSE and R² but also with their running times. Our analysis reveals that GRU, ANN–Grid–Search, ANN–Bayesian, ANN–Randomize–Search, ANN-PSO, and ANN-GA models achieve the highest performance, though this performance is inversely related to their computational cost. When the running time is included in the analysis, the GRU algorithm ranks best (RMSE: 0.945248, R²: 0.174226, runtime: 6.1), followed by ANN-GA and ANN-PSO models. These findings will help e-scooter providers plan effectively and make informed investment decisions.

As an important part of modern transportation infrastructure, high-speed rail (HSR) networks not only reduce the spatiotemporal distance between regions but also generate widespread spillover effects through mechanisms such as population mobility, technological innovation, and market expansion. Based on the city-level panel data from 2008 to 2021, this paper uses a spatial econometric model and a generalized structural equation model (GSEM) to study the spatial spillover effects of HSR networks on the three industrial agglomerations and tests the impact mechanism of HSR networks on industrial agglomeration. We find that HSR networks significantly inhibit the agglomeration of primary and secondary industries while significantly promoting that of the tertiary industry. Regional heterogeneity analysis shows that HSR networks have a negative impact on the secondary industry agglomeration in the eastern region but obviously promote the tertiary industry agglomeration, and their promotion effect on the tertiary industry is also significant in the central and western regions. The results of the mechanism test show that HSR networks significantly affect the agglomeration of the three industries through the path of population mobility, technological innovation, and market scale.

Traffic delays are a common challenge for drivers in large cities worldwide. During these delays, drivers must maintain a safe distance from nearby vehicles while avoiding collisions with pedestrians and motorcyclists. This needs frequently alternate pressing and releasing the brake and accelerator pedals, preserving low speeds. Research indicates that this repetitive action can contribute to driver’s fatigue, which is worse for manual vehicles. Other factors, such as inadequate sleep, prolonged driving, monotonous driving conditions, and heavy workloads, may also induce fatigue, further leading to ignorance of the correct seating posture, which can exacerbate the issue. Studies on driver fatigue and its prevention have been widely conducted by scholars and automotive-based industries, focusing on two subject matters: (i) driver fatigue detection systems using various technologies and (ii) fatigue prevention techniques incorporating autonomous braking systems for high-speed and long-distance driving. This paper focuses on extensively reviewing both subject matters, leading to the best proposed solution that can prevent fatigue from happening during road traffic delays at low-speed driving, as limited studies were found that can suit the traffic and social environment in developing countries, i.e., Kuala Lumpur, Malaysia. Clearly, the latter subject area focused on incorporating autonomous braking systems in the electronic control unit (ECU) of vehicles, applicable only for high-end vehicles, thus limiting accessibility. This technology can either reduce the physical effort of pedal pressing or take over the task altogether. The review will examine various causes of fatigue and the existing detection methods, compare the automatic braking solutions’ features, and propose a suitable mechanism that could benefit drivers of all types of vehicles, especially from low- to middle-end vehicles, which addresses the real needs among the affected populations with regard to road traffic delay. The outcome of this review comes in the form of a proposal for mitigating the fatigue issue from happening using a unique technique based on the research gap that is adapted to the targeted environment.