Reliable short-term freeway traffic forecasting is essential for proactive traffic management, including congestion warning, ramp metering support, and traveler information services. However, many existing forecasting models treat spatial dependencies as synchronous or weakly directional, limiting their ability to represent delayed upstream–downstream traffic propagation. This study proposes Traffic-DiMAGNet, a lightweight and interpretable lag-aware directed spatio-temporal graph neural network for real-time freeway flow forecasting. The model constructs a sparse directed sensor dependency graph by integrating physical road adjacency, training-set lead–lag traffic priors, and learnable source–target node embeddings. A lag-aware bidirectional propagation module then shifts inter-sensor messages according to estimated propagation delays, while directed random-walk normalization, directional gating, and multi-scale causal convolutions preserve asymmetric traffic semantics with low computational cost. Experiments on four Caltrans PeMS datasets show that Traffic-DiMAGNet consistently outperforms recurrent, diffusion-based, attention-based, and adaptive-graph baselines regarding MAE, RMSE, and MAPE values. The learned directed lag structures provide interpretable propagation information, and the lightweight architecture supports rolling 5 min forecasting, indicating practical potential for real-time freeway monitoring and proactive traffic management.
This study examines how often adults with a travel-limiting condition or disability use online shopping and home delivery. It uses the person file of the 2022 National Household Travel Survey (NHTS), a household-based survey, restricted to community-dwelling adults aged 18 and over. Delivery counts are heavily overdispersed, so survey-weighted negative binomial models with household-clustered standard errors are used, with delivery frequency as the outcome. Covariates are entered sequentially. Adjusting for demographics alone, adults with a travel-limiting condition do not receive significantly more deliveries; adding socioeconomic status leaves the estimate null; only after employment, driver status, and household vehicle and driver counts are added does a positive association emerge. Because those variables may be consequences rather than causes of a travel-limiting condition, this association is contingent on the model specification and does not constitute evidence of a robust unconditional difference. Within the subsample of adults who receive at least one delivery, however, the pattern is stronger and specification-stable: adults with a travel-limiting condition receive more deliveries overall and disproportionately more food, grocery, and service or personal deliveries than general goods. All estimates are cross-sectional associations and cannot establish direction or cause. The NHTS records how often deliveries occur but not why, so whether delivery compensates for inaccessible travel options remains untested and requires data on accessibility and trip substitution.
Accurate characterization of local ship motion remains challenging because conventional AIS trajectory analysis primarily relies on global or averaged motion descriptors that overlook short-term navigation dynamics. This research introduces a fractal micromovement representation framework that identifies localized motion episodes and transforms them into symbolic descriptors suitable for structural trajectory analysis. The framework combines sliding-window Higuchi fractal analysis, quantile-based episode detection, temporal aggregation, and symbolic encoding of the resulting micromovement episodes. To formalize local motion dynamics, each micro-episode is transformed into a symbolic fractal code that reflects the topological structure of the geometric complexity of the vessel’s trajectory. Application of the proposed framework to experimental trajectory data demonstrates its ability to localize intervals of elevated trajectory complexity and transform them into compact symbolic representations suitable for structural comparison. Quantitative sensitivity analysis and comparison with an independent kinematic reference further characterize the stability and operational correspondence of the fractal representation. The framework is intended for retrospective trajectory analysis, while behavioral recognition, anomaly detection, forecasting, and integration with higher-precision navigation data are considered prospective applications.
Passenger Car Equivalent (PCE) factors are widely used to convert heterogeneous traffic streams into equivalent homogeneous flow rates for the design and analysis of roads and intersections. In developing countries, mixed traffic conditions differ substantially from those in developed contexts due to variations in vehicle composition, operating characteristics, roadway parameters, and environmental conditions. Consequently, PCE values are highly context-specific and require periodic updates to accurately represent prevailing traffic conditions. However, such updates are often infrequent because conventional PCE estimation relies on extensive field data collection through time-consuming and costly traffic surveys, as well as the availability of experienced experts to conduct and validate the analyses. In Sri Lanka, the currently adopted PCE factors are more than two decades old and no longer reflect existing traffic conditions. Although several recent studies have estimated PCE values for mid-block roadway sections of various facility types (e.g., four-lane roads, two-lane roads, and freeways), no study has comprehensively addressed intersections, which are critical for signal timing and geometric design. This study aims to develop a systematic methodology for estimating intersection-specific PCE factors using drone-based video data. Traffic data were collected at selected intersections using an unmanned aerial vehicle to obtain an accurate bird’s-eye view of vehicle movements. The methodology compares the area occupancy of different vehicle categories under varying traffic compositions with that of a passenger-car-only traffic stream operating at the same average speed. Using the extracted traffic parameters, the basic headway method was applied to establish a framework for calculating PCE factors. PCE values were estimated for ten vehicle categories, and the results reveal significant deviations, particularly for three-wheelers, motorcycles, and commercial vehicles, when compared with values currently in use. A high-level comparison with studies from other developing countries in the South Asian region indicates notable differences in vehicle impacts at signalized intersections in Sri Lanka. Furthermore, the proposed methodology provides a practical, economical, and less labour-intensive approach for estimating PCE factors, enabling more frequent updates without requiring extensive field surveys or specialized expertise. Because it relies on a straightforward headway-based framework and drone-derived traffic data, the methodology can be readily adapted to different roadway facilities, including highways, rural roads, and intersections, making it suitable for application across diverse geographical regions.
Electric vehicles are emerging as a sustainable alternative to conventional transportation. However, route planning in Internet of Vehicles environments remains challenging because conventional routing algorithms based on travel distance or time do not adequately consider electric vehicle-specific constraints. Existing routing strategies often overlook the combined effects of battery energy, traffic congestion, and charging requirements, resulting in inefficient routing decisions. This paper proposes an adaptive Energy, Congestion, and Charging-Aware Routing (ECCAR) protocol that uses energy-feasibility verification, cost-based charging-station selection, and route re-optimization for electric vehicles in Internet of Vehicles environments. ECCAR integrates residual battery energy, traffic congestion, travel time, charging station availability, and charging delay into a unified routing decision framework. It first evaluates whether the remaining battery energy is sufficient to reach the destination. Otherwise, it identifies all reachable charging stations and selects the one that minimizes the routing cost rather than the nearest station. After charging, the route is recalculated using traffic and charging information obtained through V2V and V2I communications. Simulation results demonstrate that ECCAR reduces total energy consumption by up to 19.4%, travel time by up to 25.3%, and charging waiting time by up to 39.1% compared with existing routing schemes. These results demonstrate the benefits of integrating energy, traffic, and charging information for reliable electric vehicle routing in dynamic Internet of Vehicles environments.
The rapid expansion of the express delivery sector, driven by the growth of e-commerce, has increased the importance of effective complaint management as an integral component of service quality and consumer protection. Although operational service performance and complaint handling have been widely investigated, existing studies have generally assessed these dimensions separately, leaving a gap in integrated performance evaluation. This study proposes a comprehensive decision-support framework for evaluating complaint resolution performance of express delivery operators by integrating Service Failure Indicators (SFI), Complaint Resolution Indicators (CRI), an Entropy-based Composite Quality Index (CQI), and the MARCOS multi-criteria decision-making method. The proposed framework combines objective operational and complaint-related performance indicators within a unified evaluation model. The CQI is constructed by aggregating three complaint resolution indicators using objectively determined Entropy weights, while the MARCOS method is applied to rank operators according to both operational performance and complaint resolution effectiveness. The applicability of the framework is demonstrated through a case study involving five major express delivery operators in the Republic of Serbia using regulatory operational data. The results reveal meaningful differences in operator performance and show that integrating operational service failures with complaint resolution effectiveness provides a more comprehensive evaluation than evaluating either dimension independently. The proposed framework contributes to the literature by integrating operational service quality evaluation, complaint resolution performance, composite quality measurement, and multi-criteria decision-making into a unified evaluation framework. Furthermore, it provides an objective benchmarking tool that can support regulatory oversight, managerial decision-making, and continuous quality improvement in the express delivery sector.
Autonomous micromobility vehicles (AMVs) need local planning that balances task progress, safety, and pedestrian interaction quality in pedestrian-rich shared spaces. Evaluating such planners is difficult because studies vary scenarios, seeds, metrics, outputs, and aggregation rules, while single-score leaderboards hide which behaviors produce a ranking. This paper proposes a repeatable, auditable, multi-objective benchmark protocol for AMV local planning. Before comparison, it fixes the scenario set, repeated seeds, measured metrics, stored outputs, planner-interface records, and aggregation procedure. We demonstrate it with a frozen robot_sf_ll7 campaign: 47 shared-space scenarios, three evaluation seeds, and 141 scenario-seed episodes per planner in one differential-drive AMV configuration. The stress test surfaces a descriptive safety–performance separation: a Proximal Policy Optimization (PPO)-family profile reaches higher observed mean task success than a classical reciprocal-avoidance baseline, while that baseline keeps lower collision exposure. Absolute success stays low for both, with most scenarios unsolved by either. Because this learned policy was trained on a superset of the evaluation scenarios, its higher success reflects behavior on the benchmark set, not held-out generalization—an overlap the protocol records per planner rather than hiding in one score. The finding is bounded to these configured pipelines, not a universal planner-family ranking. The contribution is an auditable comparison framework tracing results from the scenario matrix and fixed seeds to episode records, aggregate reports, and manifests.
Severe environmental pollution and complex weather changes significantly hinder the effectiveness of vehicle traffic flow statistics and traffic monitoring technologies. Therefore, achieving fast and accurate vehicle detection in complex environments has become one of the key tasks in the new era. This paper proposes a vehicle detection method for complex environments based on YOLOv11, named YOLO-FTG. First, the neck network of the YOLOv11 baseline model is improved by adding a P2 detection layer and the corresponding detection head. Second, a Spatial-Frequency Hybrid Convolution (SFHC) module is designed. Third, a Global-Local Adaptive Module (GLAM) is proposed. To verify the effectiveness of the proposed model, experiments were conducted on three self-constructed datasets and the public BDD100K dataset. The experimental results demonstrate that compared with existing methods, the YOLO-FTG model achieves higher accuracy, with mAP50 scores of 75.3%, 98.71%, 51.00%, and 64.59% on the four datasets, respectively. These scores represent improvements of 3.24%, 0.34%, 5.64%, and 3.51% over the baseline model, respectively. While maintaining real-time inference speed, these results indicate the effectiveness and robustness of the proposed model in complex environments.
Mobility hubs promise to reduce car dependence and make multimodal travel work in practice, yet behavioural evidence remains limited when hub improvements coexist with easier car access. This article examines the tension at Rome Trastevere, an urban rail node that gradually acquires mobility-hub functions while facing improved parking access near Piazza della Radio. The empirical analysis combines a pilot survey of 83 users with an on-site stated preference survey of 204 valid respondents. The stated preference instrument uses a route-based feasible-choice design with nine choice sets per experiment: respondents evaluate alternatives among bikes, walking, e-scooters, e-mopeds, public transport, private cars, and shared cars under variations in travel time, travel cost, and search time. The paper estimates a multinomial logit model in Apollo and uses sample enumeration, supported by Monte Carlo simulation, to assess four parking and shared-mobility scenarios and produce confidence intervals around predicted probabilities. Results show that users respond to time, monetary cost, and search friction in coherent and policy-relevant ways. Setting the car parking search time to zero increases predicted car probability only marginally, by about 0.9% relative to the baseline. By contrast, a EUR 1/h increase in parking cost reduces predicted car probability by about 14.7%, while a EUR 1.5/h increase reduces it by about 22.4%. A coordinated scenario combining higher parking cost and lower shared-mode search time produces the lowest predicted car probability and strengthens e-scooter and e-moped alternatives, while public transport remains the dominant option. Findings indicate that parking pricing steers behaviour more clearly than parking convenience destabilizes it in the tested range. The paper shows that mobility-hub performance depends on coordinated access management, including parking regulation, shared-service reliability, and legible multimodal transfer.
This study examines the determinants of Generation Z’s purchase intention for environmentally friendly public transportation in Indonesia, a mode of transport that continues to experience suboptimal ridership. We investigate the effect of consumers’ awareness of green product attributes and operational processes, as well as their exposure to green promotions, on green brand sustainability and, subsequently, on green purchase intention. The study involved 400 Generation Z respondents aged 19 years or older residing in Jakarta, Indonesia. Data were analyzed using Structural Equation Modeling (SEM) with LISREL11.0. The results show that green product awareness, green process awareness, and green promotion exposure significantly influence green brand sustainability, explaining 69 percent of its variance, with green product awareness being the most dominant predictor. Furthermore, green purchase intention is significantly shaped by these antecedents and green brand sustainability, collectively explaining 84 percent of its variance. The findings demonstrate that strengthening environmental awareness across product attributes, operational processes, and promotional communication can substantially increase Generation Z’s intention to use sustainable public transportation. This study extends the research on green consumer behavior and provides strategic insights to strengthen sustainability in the public transportation sector.
This paper proposes a norm-based admissibility criterion formulated in the frequency domain for evaluating whether translational and rotational motion amplitudes of a ship moored at a quay remain within operational limits prescribed by port authorities. The approach is built on a linear six-degree-of-freedom model that includes hydrodynamic added-mass and radiation-damping effects, wave excitation forces, and aerodynamic wind loads, as well as linearized reactions of mooring lines and quay fenders, including an equivalent viscous representation of hull–fender friction. Instead of explicitly inverting the full system matrix to compute the complete response, the admissibility assessment is derived from row-wise norm bounds of the frequency-domain system, yielding a computationally efficient admissibility criterion for compliance with motion limits. The criterion naturally enables a port-oriented decision index and an operational safety margin that can be evaluated for each degree of freedom and used to compare alternative mooring arrangements. Numerical verification is performed for a bulk carrier under storm wave excitation and different loading conditions, demonstrating the sensitivity of admissibility to mooring geometry and pretension. The results confirm that the proposed criterion provides a practical engineering tool for rapid go/no-go decisions regarding cargo operations and supports the selection of mooring arrangements that improve operational robustness under adverse environmental loading conditions. In addition, a Monte Carlo-based uncertainty analysis is performed to evaluate the robustness of the proposed admissibility criterion with variable mooring stiffness and damping parameters. The proposed criterion is intended as a rapid engineering screening tool to complement conventional frequency-domain response analysis.
This study investigates the influence of spatial heterogeneity on urban transport demand forecasting. A two-stage framework combining cluster analysis and machine learning was applied to origin-destination trip data generated by a calibrated transport model of a large city. Origin-destination pairs were grouped according to travel conditions using K-Means clustering based on private transport trip length and public transport travel time. The identified clusters were subsequently analyzed with respect to travel costs, modal split, spatial distribution, and transport demand characteristics. Machine learning models were then developed to predict transport demand for public transport, private transport, and walking. Two forecasting strategies were compared: a unified model trained on the entire dataset and cluster-specific models trained separately for each identified cluster. The results revealed significant spatial heterogeneity in travel conditions and transport demand structure. Cluster-specific machine-learning models reduced prediction errors for public and private transport demand, while the magnitude of improvement varied across travel modes and evaluation metrics. The findings demonstrate that accounting for spatial heterogeneity influences transport demand prediction performance, with the greatest improvements observed for public and private transport demand, whereas the effect was less pronounced for walking demand.
The forecasting of road surface conditions is a pivotal component for intelligent transportation systems, in terms of supporting maintenance planning, safety and mobility management. The increasing availability of large-scale monitoring data, collected from passenger vehicle fleets, enables the development of data-driven forecasting approaches. However, systematic comparisons between classical time-series models and machine-learning methods in this context remain limited. The proposed benchmarking framework evaluates direct road surface roughness forecasts at 1-, 7-, 14-, 30-, and 90-day horizons using multi-year vehicle-derived data collected across heterogeneous road segments. Daily roughness indicators are derived from raw measurements and modeled following a consistent, segment-wise experimental protocol. The proposed analysis involves the evaluation of multiple machine-learning regressors including Ridge, Random Forest and Gradient Boosting which are trained on lagged observations and rolling statistics. Performance of the models is assessed using two error metrics: unweighted and uncertainty-aware weighted. Findings indicate significant variations in predictive accuracy and robustness across models and segments, emphasizing the influence of feature-based learning strategies and data-quality weighting. The research provides a scalable and transparent methodology for evaluating forecasting models on vehicle-based road monitoring data, contributing practical guidance for the deployment of artificial intelligence in Intelligent Transport Systems (ITSs).
Digitalization is transforming urban mobility and logistics, changing behaviors and the way demand is anticipated and managed. This paper frames both research and practice in this area. Through a systematic review of reviews in Scopus and Web of Science, following PRISMA 2020, a corpus of 21 documents was compiled. The analysis organizes findings into five interrelated dimensions: technology; operation and service design; society, user, and equity; institution, regulation, and governance; and economics and scalability. These are interpreted through two cross-cutting axes: behavioral change among users and operators, and the anticipation and management of demand supported by big data and predictive models. By integrating urban mobility and logistics, usually analyzed separately, this study compares barriers and enablers. The two axes help identify gaps: limited coverage of medium-sized cities, low-density environments, and developing countries; fragmented treatment of user diversity; weak integration between mobility and logistics data; and the lag between the sophistication of predictive models and the institutional capacity to incorporate them into planning. This paper proposes an integrative framework for analyzing the digitalization of urban mobility and logistics as part of a single urban transition.
The safe and uninterrupted operation of the ship’s main engine is critical for maritime transportation. The shutdown mechanism, part of the main engine protection systems, prevents serious damage by automatically stopping the engine in critical situations such as low lubrication oil pressure, overspeed, high bearing temperature, and cooling system failures. However, identifying the faults that trigger the shutdown system and evaluating their risk levels is crucial for improving system reliability. In this study, shutdown events that may occur in a two-stroke low-speed marine diesel main engine were investigated using Fuzzy Fault Tree Analysis (FFTA). The shutdown event was defined as the peak event, and a total of 34 baseline events were modelled under five main branches: low lubrication oil pressure, overspeed, high thrust bearing temperature, abnormal jacket coolant inlet condition, and crankcase/cylinder oil mist formation. Fuzzy assessments based on expert opinions were defuzzified and converted into probability values and used in fault tree calculations. The results showed that the shutdown risk is largely affected by failures originating from the jacket coolant system and the lubrication oil system. Specifically, lubrication oil filter clogging and contamination/blockage in the coolant line were identified as the most critical risk factors. The findings significantly contribute to prioritizing maintenance and condition-monitoring activities aimed at improving the ship’s main engine reliability through a risk-based approach.
To mitigate driving risks from brake failure on long and steep downhill sections, this study designs three deployment schemes for radar–video fusion devices: a baseline scenario with no coverage, a scenario with partial coverage in high-risk areas, and a scenario with full coverage. Corresponding information service strategies are delivered via Human–Machine Interfaces (HMIs), forming an integrated active prevention and control framework from risk perception to preventive action. Driving simulation experiments focusing on the car-following process were conducted to collect vehicle operational data and extract characteristic indicators based on the Wiedemann model. A Generalized Linear Mixed Model was employed to comprehensively examine the effects of HMIs on car-following behavior to identify the optimal active prevention strategy. Results show that drivers exhibit greater caution under the partial coverage scheme, with time headway increasing by 47.63% compared to the scheme with no radar–video fusion devices to ensure safety. Under full coverage conditions, drivers can obtain real-time information about the leading vehicle’s status and the distance between the two vehicles in key risk sections. Drivers choose to follow the leading vehicle, balancing both safety in car-following and efficiency on long and steep downhill sections. As the level of accompanying services improves, drivers engage in self-regulation to avoid rear-end collisions. Particularly under the scheme with full coverage of radar–video fusion devices, the standing distance significantly increases by 219.37% compared to the partial coverage condition. Drivers demonstrate optimal vehicle control capabilities. Furthermore, there is an interaction effect between the accompanying service strategy and drivers’ attributes on car-following behaviors. Under different schemes, more experienced drivers exhibit a certain degree of aggressiveness, providing a basis for the targeted design of information services for different types of drivers. The findings support the deployment and application of risk perception and prevention devices on long and steep downhill sections, which can effectively enhance the comprehensive safety of such special roads in the connected vehicle environment.
Work zones reduce roadway capacity and create unstable merging, queue spillback, and stop-and-go conditions that degrade traffic operations while elevating crash risk. Conventional fixed-time, actuated, and adaptive controllers are poorly suited to these non-stationary conditions, and most reinforcement learning approaches optimize mobility while treating safety only as a post hoc evaluation measure. This study develops a safety-aware Deep Q-Network framework for adaptive signal control at intersections operating near work zone activity areas. Merge conflict risk, upstream spillback propagation, and stop-and-go instability are embedded directly into both the state representation and the reward formulation, alongside operational objectives. A merge-conflict model based on relative spacing, relative speed, and acceleration characterizes unsafe interactions in the merge region, and a Pareto-based procedure samples reward-weight vectors to identify non-dominated policies. The framework was evaluated in a SUMO microscopic simulation of a signalized intersection under lane closure. Relative to default fixed-time control, the selected policy increased throughput by 24.6–37.3% across vehicle classes (p < 0.001; Cohen’s d = 0.53–1.29), with the largest gains for trucks and buses, and reduced maximum queue length by 39.1% and spillback distance by 45.8%. The findings show that a single controller trained with surrogate safety indicators as learning objectives can improve operational performance while reducing safety-critical instability in work zones.
This study proposes a rigorous optimization framework for the design of traffic counting station locations in large-scale highway networks, with specific application to Thailand’s national highway system. A mixed-integer linear programming (MILP) model is developed to determine the optimal sensor placement under budget-constrained scenarios while explicitly incorporating existing infrastructure. The model aims to maximize origin–destination (OD) flow observability and minimize estimation error, measured by the percentage of OD flows intercepted and root mean square error (RMSE). The proposed framework is validated using a real-world network. The results demonstrate that the optimized design significantly outperforms conventional approaches, including random and high-flow-based selection methods, achieving over 70% reduction in estimation error and 93% of OD flows intercepted with a feasible number of stations. Furthermore, the statistical representativeness of the selected locations is validated across spatial, functional, and traffic characteristics and traffic measurement errors. The findings provide a scalable and cost-effective decision-support tool for transport authorities in developing countries seeking to modernize transportation planning, traffic management, and infrastructure development under limited resources.
Railway transport is increasingly recognized as a key pillar of sustainable mobility, offering a low-carbon and energy-efficient alternative to road and air transport and playing a critical role in achieving climate objectives, regional connectivity, and sustainable tourism development. Despite extensive research on service quality, sustainability, and tourism, their interrelationship within the railway sector remains insufficiently explored. This study aims to systematically analyze the intersection of quality management systems (QMS), sustainability, and tourism in passenger railway transport and to identify structural gaps that hinder their integration. A systematic literature review was conducted following the PRISMA methodology, resulting in a final sample of 37 studies. The findings reveal a significant research gap, particularly the absence of integrated and empirically supported QMS frameworks linking passenger satisfaction with sustainability and tourism objectives. Quality-management-oriented constructs appear in 48.6% of the analyzed studies, sustainability in 32.4%, and tourism in 24.3%, while none demonstrate full integration of all three dimensions. The study contributes by providing a conceptual basis for future research on the integration of operational quality management, environmental performance, and passenger-oriented service quality in railway systems.
This article conducts a thorough comparative analysis of public transport systems in Curitiba and Lisbon, focusing on cost-efficiency and structural performance from the user’s viewpoint. Curitiba is noted for pioneering the BRT model in the 1970s, while Lisbon is evolving towards a multimodal system with substantial investments in integration and user-centric policies. Employing a case study methodology and mixed analytical approaches, the analysis examines governance structures, network architecture, financing mechanisms, and service quality indicators. The findings indicate that although Curitiba imposes a similar or higher fare burden relative to user incomes, it offers significantly lower service value across various dimensions, including modal diversity and infrastructure quality. In contrast, Lisbon’s integrated governance model for bus and tram networks proves effective in enhancing accessibility and sustainability, despite some coordination issues with centrally governed transport networks. This study contributes to the international discourse on the limitations of single-modal transport systems and highlights the necessity of institutional integration, long-term investment, and adaptive governance frameworks for urban mobility transformation in the 21st century.
