This study proposes a novel approach to driver drowsiness detection using the Video Vision Transformer (ViViT) model, which captures both spatial and temporal dynamics simultaneously to analyze eye conditions and head movements. The National Tsing Hua University Driver Drowsiness Detection (NTHU-DDD) dataset, which consists of 36,000 annotated video clips, was utilized for both training and evaluation. The ViViT model is compared to traditional Convolutional Neural Network (CNN) and Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) models, demonstrating superior performance with 96.2% accuracy and 95.9% F1-Score, while maintaining a 28.9 ms/frame inference time suitable for real-time deployment. The ablation study indicates that integrating spatial and temporal attention yields a notable improvement in model accuracy. Furthermore, positional encoding proves essential in preserving spatial coherence within video-based inputs. The model’s resilience was tested across a range of challenging conditions including low-light settings, partial occlusions, and drastic head movements and it consistently maintained reliable performance. With a compact footprint of just 89 MB, the ViViT model has been fine-tuned for deployment on embedded platforms such as the Jetson Nano, making it well-suited for edge AI applications. These findings highlight ViViT’s promise as a practical and high-performing solution for real-time driver drowsiness detection in real-world scenarios.

An unguarded railway level crossing in Padang, Indonesia, presents a critical safety risk. This study investigates how community group size, social capital, and duration of residence in the neighborhood interact to influence collective action for shared safety. The collective action takes the form of conscious participation in an initiative to finance railway crossing guards, provided by and for the community. Using two-stage probit regression, the analysis uses the duration of residence in the neighborhood as an instrumental variable for social capital to address potential endogeneity. Not necessarily, the longer residents live in the neighborhood, the greater the willingness to participate. It is possible that living longer in an area leads to a decline in social capital due to economic stagnation and social fragmentation. Especially in disadvantaged neighborhoods. These dynamics illustrate the complex interactions between the duration of residence, group size, and social capital. These interactions will lead to a wide variety of responses in maintaining grassroots safety efforts.

Automated Guided Vehicles (AGVs) are increasingly used in industrial and logistics operations for material handling, offering benefits such as reduced human error, improved efficiency, and lower operational costs. This study presents the design and implementation of a real-time intelligent management system for Forklift AGVs based on deep learning techniques. The core of the system is an optimized version of YOLOv3, termed YOLOX, enhanced with Adaptive Spatial Feature Fusion (ASFF) and advanced data augmentation strategies. The ASFF module employs spatially adaptive weights (α, β, γ) to dynamically integrate multi-scale features across the Feature Pyramid Network, improving the detection of small, occluded, and overlapping objects. The system is trained on a combined Pascal VOC dataset using mix-up and label smoothing to enhance generalization and model robustness. It is deployed on embedded hardware, including Raspberry Pi 4, enabling real-time processing of visual data and sensor inputs under various lighting and environmental conditions. Evaluation results indicate that the model achieves a high mean Average Precision (mAP) of 94.17%, with real-time confidence scores reaching 98.1% in natural lighting and 94.3% in dim conditions. The system effectively detects and classifies a wide range of objects—including static, dynamic, small, distant, and partially occluded—in complex scenes. The proposed solution demonstrates robust real-time performance and adaptability, making it suitable for deployment in resource-constrained environments. It offers a scalable and intelligent framework for autonomous AGV navigation, contributing to safer and more efficient material transportation in real-world applications.

This research aims to analyze the existing vehicle speeds on a primary arterial road and to implement speed management on road segments to improve road safety. This study utilizes secondary data collected from local Police accident reports over the four preceding years (2019-2022). A road inventory survey was conducted to determine road characteristics. Traffic count surveys and spot speed surveys were used to obtain traffic characteristics. Existing speeds were analyzed using the 85th percentile speed, which reflects the operating speed of the majority of drivers. Subsequently, the 85th percentile speed was benchmarked against Indonesia’s Minister of Transportation Regulation No. PM 111/2015. A Chi-square analysis was used to test the influence of vehicle speed on the fatality rate of traffic accident victims. The results of the Chi-square test indicated a significant relationship between vehicle speed and the fatality rate of traffic accident victims on the Jalan Raya Anyer. To improve road safety, speed control was implemented through the installation of signs and markings, particularly in accident-prone areas. The installation of speed limit signs is proposed in areas with educational, commercial, and residential activities. To enhance effectiveness, speed limit signs are installed repeatedly at a distance of 35 m between signs. The results of this research can contribute to the development of more effective policies and strategies to improve traffic safety on primary arterial roads, particularly in terms of speed management.

Bengaluru, a city renowned for its rapid urbanization and booming population, faces severe traffic congestion that threatens road safety, increases environmental pollution, and disrupts the daily lives of its residents. The persistent delays at traffic lights and extended commute times underscore the urgent need for effective solutions. In response to these challenges, this study focuses on employing advanced machine learning techniques, specifically, convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and support vector regressions (SVRs) to analyze and predict traffic congestion patterns within the city. By leveraging the strengths of CNNs, the system is designed not only to provide accurate congestion detection across multiple locations but also to offer optimal routing recommendations to road users, thereby potentially easing traffic flows. To comprehensively evaluate the proposed approach, its performance is benchmarked against LSTM and SVR models using key performance metrics such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and the R² coefficient. These metrics ensure a robust assessment of predictive accuracy and model effectiveness.

The increasing desire for people to own personal cars, combined with their reluctance to use public transportation, has led to traffic jams and delays in emergency vehicle arrivals. Traffic lights in densely populated cities pose a significant challenge because they rely on fixed or variable timings, yet are not particularly effective. As a result, they can worsen congestion or cause traffic jams instead of alleviating it. For example, a city like Baghdad faces severe traffic congestion, requiring intervention from traffic police. Additionally, there is no specific system in place for emergency vehicle passage, and public transportation remains ineffective, as people are hesitant to use buses due to longer congestion times and the difficulty in navigating, which is exacerbated by their larger size compared to private small cars. Unlike previous YOLO-based systems, our system integrates emergency vehicle and public transport buses prioritization. It adjusts timing based on vehicle type, number, and estimated speed, showing a 31.11% improvement in flow efficiency and reducing queue delays by 21.64% compared to fixed-time signal systems. The improved algorithm can recognize all four vehicle classes (fire trucks, ambulances, public transport buses, and cars) with an accuracy of 85-99%, depending on vehicle density and complex lighting conditions.

infrastructure and trade connectivity across Asia. This study assesses the impact of BRI-funded transport infrastructure projects on agricultural trade efficiency in Southeast Asia, focusing on key projects such as the China-Laos Railway and Malaysia’s East Coast Rail Link (ECRL). The research employs a mixed-methods approach combining trade flow analysis, policy document review, and semi-structured stakeholder interviews. The findings reveal that transport costs for agricultural products decrease by up to 50%, while transit times are halved, particularly benefiting perishable goods such as fruits and vegetables. Export volumes of staples such as rice and cassava increase substantially, with durian exports to China reaching USD 3 billion annually. Despite these achievements, challenges remain, including limited access for smallholder farmers, insufficient rural infrastructure, and logistical bottlenecks in cold-chain systems. By integrating recent data and insights, this study underscores the need for targeted policies, such as harmonised trade regulations and investments in rural connectivity, to maximise the equitable and sustainable benefits of BRI infrastructure for agricultural trade.

Since July 2022, Indonesia's Electronic Customs Declaration (ECD) system has replaced traditional paper-based forms for international arrivals. Although intended to enhance efficiency and user convenience, the system has generated considerable dissatisfaction, with 83.14% of recorded complaints citing technical or usability challenges. This study examines the determinants of passenger satisfaction with the ECD system through an integrated framework that combines the Expectation-Disconfirmation Model (EDM), the Unified Theory of Acceptance and Use of Technology (UTAUT), and the DeLone & McLean Information Systems Success Model. A cross-sectional survey of 207 Indonesian international passengers at Soekarno-Hatta International Airport was analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM), suitable for evaluating latent constructs and mediation effects in complex models. The results indicate that effort expectancy, drawn from UTAUT, significantly influences perceived system quality and Disconfirmation, both of which serve as critical mediators of user satisfaction. System Quality, based on usability, reliability, and interface design, exerts an indirect effect through Disconfirmation, as conceptualized in EDM and DeLone and McLean's framework. Collectively, the variables explain 86.3% of the variance in satisfaction, with all key paths statistically significant (p < 0.001). These findings underscore the importance of expectation alignment, ease of use, and perceived system quality in shaping satisfaction with digital public services and provide practical insights for user-centered design and implementation of government technologies.

The multimodal transport system in India poses severe accessibility issues for wheelchair users and other differently-abled riders. While there are policies designed to address these issues, gaps still exist in applied research focusing on the complete understanding and evaluation of these challenges, particularly pertaining to the Indian context. This paper aims to analyze literature on the most pertinent issues related to physical and infrastructural restrictions, including the boundless social perception problems overriding empathic understanding people have towards aiding others. The results of the study pointed out the widespread inadequate infrastructure—a lack of ramps, elevators, tactile guides, accompanying accessible last-mile services, policy gaps, enforcement inequality, and unique problems regarding ride-sharing services like algorithm discrimination and insufficient vans equipped with lifts. Additional social perception issues based on legislation, technology, and grassroots efforts also challenged the existing government policies. Through addressing these problems, this study introduces an advocacy strategy aimed at revamping policies, infrastructure, education, technology, and the role of government and civil society for a supportive environment involving everyone at the community level. The primary significance of this research is the development and proposed concepts concerning the multimodal transport policy from the perspective of methods used to implement policies in relation to the use of Indian systems of evaluation of accessibility in transportation. This tool aims to empower policymakers, technologists, and business stakeholders to systematically identify, measure, and address the accessibility needs of differently-abled passengers, thereby fostering a more inclusive and equitable transportation system in India.

This research was conducted at the Fakfak Port in the West Papua region, which serves as a collection port and logistics distribution center. This study aims to determine the extent of the readiness of the Fakfak Port in terms of its potential strengths, weaknesses, and opportunities that could support the Fakfak Port as a consolidation port. The data obtained are both qualitative and quantitative and are processed using the Strengths, Weaknesses, Opportunities, Threats (SWOT) analysis method. Furthermore, it is arranged with the Internal External (IE) matrix to determine the company's position. The study results show that the greatest strength, namely the operational performance and utility of the Fakfak Port facilities, has improved from year to year with a significance level of 4.0 and a weight of 3.0. However, the weakness of the Fakfak Port is the lack of separation between passenger and cargo ports, with a significance level of 3 and a weight of 1.8. Opportunities that can be utilized are the supporting nodes for industrial and trade activities in the Fakfak Regency, with a significance level of 3.5 and a weight of 3.2. The biggest threat to Fakfak Port is the faceline of Pier I (water depth in front of Pier I), which is only ± 3 m. The IE results show that the Fakfak Port has the opportunity to be used as a consolidation port based on the factors.

The study aims to evaluate sustainable traffic management strategies for congested intersections in medium-sized Iraqi cities, with a focus on Al-Sa’a Intersection and Al-Jari Street in Hit City. These nodes face severe traffic congestion, delays, and infrastructure limitations that compromise urban mobility and sustainability. A multi-criteria evaluation (MCE) framework was employed to analyze three categories of interventions—engineering, planning, and administrative—based on five weighted criteria: traffic efficiency (40%), delay reduction (25%), cost (20%), environmental impact (10%), and social acceptance (5%). The methodology combined field data collection (traffic counts, travel time, and delays), GIS-based spatial analysis, and stakeholder consultation to prioritize solutions and evaluate performance. The findings indicated that all proposed solutions improved traffic performance, but varied in scope and impact. Engineering solutions, such as street widening and grade separation, reduced congestion by up to 40%. Planning measures, including public transport enhancement and alternative routes, scored the highest (8.2/10) due to their long-term sustainability and balanced environmental impact. Administrative actions—optimized signal timing and truck regulation—offered low-cost, short-term improvements. The study demonstrates the value of integrated, GIS-supported, multi-criteria approaches in diagnosing and addressing urban traffic challenges in secondary cities. A phased implementation strategy is recommended: initiate with administrative measures, transition to planning-based interventions, and apply engineering upgrades where necessary. The framework can support future transport planning in similar urban contexts across Iraq.

Global urbanisation is evident in Sub-Saharan Africa, especially Nigeria, where the population has steadily increased by 3.2% annually. This increment necessitates the adoption of sustainable public transportation, with rail transport leading the advancement. However, train terminals are fraught with complex and poorly implemented approaches to pedestrian circulation. This study evaluated the implementation of pedestrian circulation strategies within three existing train terminals in Lagos, Nigeria, aimed at determining their influence on optimal user experience. The research method employed in this study is a mixed-method approach, which entailed the distribution of survey questionnaires to 60 respondents. Thereafter, descriptive statistics were thoroughly carried out using the IBM Statistical Package for Social Sciences (SPSS) version 27. The results show that the pedestrian circulation strategy that influenced user experience the most within the selected train terminals was the connection of corridors and lobbies with other facilities. Therefore, it is recommended that horizontal pedestrian circulation strategies should be appropriately spatially planned and dimensioned to accommodate high pedestrian traffic scenarios within train terminals.

The aviation industry is experiencing a rapid digital transformation driven by globalization, technological advancements, and evolving customer expectations. Among these technologies, AI-based chatbots have emerged as a powerful tool to streamline operations, enhance customer service, and support internal business functions. However, their adoption in air ticket reservation services is still in its early stages. This study aims to provide innovative insights into understanding the factors that determine the adoption of AI-based chatbots for air ticket reservations from the organization’s perspective. The study introduces two new constructs, diversity and sensibility, and conceptually integrates the “Technology Organization Environment” theory and the “Diffusion of Innovation” theory. Data from 154 respondents were modeled using PLS-SEM, suitable for models with many variables and small sample sizes. The finding reveals that the organization's technical capability is a key factor influencing the adoption. Diversity, referring to the chatbot’s multifunctionality, promotes broader acceptance. Moreover, the impact of sensibility on adoption intention posits that a user-friendly design of the chatbot that enhances the “look” and provides a sense of human touch significantly increases the adoption intention. The relative advantage of AI-based chatbots on adoption illustrates that among all other ticket reservation channels, they prove to be the most efficient and profitable. Also, the complexity and government involvement were identified as relevant predictors of adoption. This study provides valuable insights for organizations and stakeholders and offers both theoretical and practical implications. The study concludes with limitations and proposes directions for future research.

This study developed and evaluated a simulator-based instructional model to enhance cadet pilot competencies in safety awareness, decision making, and flight performance. Using the ADDIE instructional design model, the research involved 60 cadets from the Indonesian Aviation Academy Banyuwangi. A pre-test/post-test design was employed, with data analyzed using paired sample t-tests. The SIM-FLIGHT model was implemented through structured modules, scenario-based assessments, and instructor guidelines using the Redbird FMX1000 flight simulator. The results showed significant improvements in all competency areas: safety awareness increased by 24.70 points, decision making by 24.62 points, and flight performance by 54.22 points (all p < 0.001). Importance-Performance Analysis (IPA) confirmed that all competencies fell into the high-importance/high-performance quadrant, indicating alignment between instructional goals and outcomes. These findings highlight the effectiveness of integrating simulation with structured instructional models. Practically, the model offers a scalable framework for aviation institutions to enhance competency-based training through data-driven, simulation-integrated approaches.

Advances in wireless communication and sensor technologies have enabled vehicle-to-vehicle (V2V) systems that enhance road safety and traffic efficiency. The objective of this study is to develop and evaluate a multi-agent V2V communication framework that enables cooperative driving, allowing autonomous vehicles to make real-time, informed decisions in complex traffic scenarios. The proposed system is implemented using the JADE multi-agent platform and incorporates reinforcement learning and cooperative decision-making strategies. Each vehicle is represented by a Generic Car Agent (GCA) with integrated sub-agents responsible for driver modeling, information integration, knowledge management, and active interface functions. Remote Car Agents (RCA) and Traffic Control Agents (TCA) facilitate communication across vehicles and traffic networks, enabling coordinated maneuvers such as lane changes and platooning. The framework is evaluated using real-world traffic data collected from urban and highway roads in Jordan, across five challenging driving scenarios. Simulation results show improved traffic flow, reduced collision risk, and enhanced fuel efficiency. The system is cost-effective, leveraging existing onboard sensors and standard wireless technologies, demonstrating practical potential for scalable deployment in intelligent transportation systems.