IntroductionBiking-both shared and non-shared-has become a vital component of sustainable urban mobility across African cities like Kigali. Despite this progress, empirical research and demand modeling of biking behavior remain limited. This study predicts and explains biking behavior in Kigali by integrating advanced ensemble machine learning (ML) techniques with structural equation modeling (SEM). MethodsA dataset of 6,386 observations was compiled by merging survey responses on biking with secondary data on weather and air quality. Both traditional statistical and advanced ensemble models were developed for comparison. The dataset was partitioned into training (70%) and testing (30%) subsets, with correlation-based and model-based feature selection applied. SEM examined latent constructs representing spatial, social-demographics, temporal, environmental, and attitudinal factors. ResultsEnsemble ML models substantially outperformed traditional approaches, with random forest and XGB classifiers achieving the highest predictive performance. The SEM demonstrated good model fit and explained the variance in biking frequency. Perceived station accessibility emerged as the strongest determinant of biking behavior, while temporal and environmental factors indirectly influenced demand patterns. DiscussionThe combination of ML and SEM has revealed a coexistence of accurate prediction and behavioral insight. Accessibility emerged as central to biking uptake, highlighting the potential of station placement and spatial equity. Indirect effects of temporal and environmental conditions highlight the impact of user perceptions in shaping biking demand. ConclusionIntegrating ensemble ML and SEM provides predictive robustness and behavioral insight. The findings highlight that improving spatial accessibility and adopting adaptive urban planning strategies enhance sustainable biking uptake.
Introduction/Objective The need for sustainable logistics in the bottled water distribution sector is growing rapidly. This study addresses this issue by developing a multi-objective optimization model that creates a balance between travel distance and carbon dioxide emissions. The aim is to create and design efficient delivery routes while utilizing intelligent vehicle allocation to reduce environmental impact. Methods A mixed-integer linear programming model is proposed, integrated with K-means clustering and real-time traffic data imported from Google Maps API. Two different case studies were examined: one employed a heterogeneous fleet consisting of large, medium, and small-sized vehicles, and the other one employed a homogeneous fleet consisting of small vehicles with various emission rates. The two cases were analyzed according to three weighing scenarios: equal priority (50%–50%), distance-prioritized (80%–20%), and emissions-prioritized (20%–80%). Results In the first case, the model selected the same fleet composition in all scenarios. The total objective function values were 467.30, 546.15, and 388.42. In the second case, the model assigned the heaviest loads and longest routes to lower-emission vehicles, resulting in objective values of 65.34, 99.49, and 31.20, with stable route assignments and optimized load utilization. Discussion The model showed strong adaptability across fleet structures and operational priorities. It was able to reduce emissions without compromising operational efficiency, which shows its practical value for logistics planning. The utilization of real-time traffic and emission-based routing enhances its environmental applicability further. Conclusion The proposed approach supports sustainable route planning and fleet utilization. According to the various scenarios and vehicle configurations used to test the robustness of the model, it can be used on a broader scale in logistic systems with environmental constraints.
Introduction The study of car adoption examines the processes and determinants that shape individuals’ decisions to purchase vehicles. These decisions result from a complex interaction between personal characteristics and contextual factors, including technological, social, and environmental considerations. Although academic interest in this topic has increased substantially, the literature remains fragmented, revealing theoretical and empirical gaps that limit a comprehensive understanding of car purchase behavior. Methods A systematic literature review was conducted in accordance with the PRISMA 2020 guidelines. The search strategy was applied to the Scopus and Web of Science databases, focusing on peer-reviewed empirical studies addressing vehicle purchase and adoption decisions. Eligible studies were screened, assessed, and analyzed using descriptive and thematic synthesis techniques. Results The findings indicate that questionnaires and surveys are the predominant data collection instruments. The highest research output originates from the United States, China, and India. The most frequently analyzed populations include car buyers, electric vehicle consumers, drivers, and university students. The most commonly employed theoretical frameworks are the Theory of Planned Behavior, the Technology Acceptance Model, and the Protection Motivation Theory. Key variables repeatedly examined include car purchase intention, age, perceived behavioral control, attitudes, and social norms. Discussion The results reveal a strong concentration on a limited set of theories and variables, suggesting a need to expand analytical perspectives. This theoretical convergence may limit the explanatory power of existing models and overlook emerging factors such as sustainability concerns, technological transitions, and market-level heterogeneity. Conclusion This review highlights the need to adopt more integrative and diverse theoretical approaches to advance understanding of car adoption. Future research should incorporate broader conceptual frameworks and contextual variables to better capture the complexity of vehicle purchase decisions.
Introduction Advanced Traveler Information Systems (ATIS) provide real-time route and traffic data to drivers, allowing them to adjust travel based on shared insights. This study examines the factors influencing ATIS adoption and employs a Dynamic Traffic Assignment (DTA) model to evaluate varying levels of information access within the Bucaramanga Metropolitan Area. Methods To achieve this, surveys were designed to identify the variables affecting drivers' route choice behavior. Based on these insights, key variables for ATIS acceptance were defined, and seven DTA scenarios were simulated under specific assumptions of guidance acceptance or rejection. Results The results indicate that drivers primarily consult ATIS to avoid congestion, save time, and navigate unfamiliar routes. Consequently, utilizing this real-time information increases travel speeds and reduces incident-related congestion. However, a critical efficiency threshold emerged: performance improvements are strictly limited to a 0-30% acceptance rate locally and 0-50% network-wide. When over 50% of drivers act on ATIS guidance, overall network performance declines. Discussion This finding highlights the necessity of careful monitoring and strategic traffic management to prevent secondary congestion on alternative routes. Conclusion Ultimately, while providing real-time information effectively mitigates traffic incidents, the benefits of ATIS are non-linear. Maximizing urban network efficiency requires strategically managing information distribution to keep user acceptance within these optimal local and global thresholds.
Introduction Many large cities, particularly in Latin America, have adopted Bus Rapid Transit (BRT) systems to reduce emissions from individual road transport, yet their impacts on air quality are rarely evaluated. This study aims to estimate fuel consumption and emissions from the BRT fleet in Guayaquil and to explore several decarbonization scenarios. Methods The study applies macroscale Tier 1, Tier 2, and Tier 3 methodologies from the European Environment Agency to estimate fuel consumption and emissions of the Guayaquil BRT fleet. Several “what if” decarbonization scenarios are assessed, including upgrading emission-control technologies, switching to low-sulfur fuels and higher biodiesel blends, and incorporating air-conditioning systems. The analysis also accounts for refrigerant leakage and key upstream CO 2 emissions from fuel desulfurization and biodiesel production. Results The results reveal substantial uncertainty in pollutant inventories. For example, NO X emissions range from 433 to 724 tons per year, depending on the applied methodology. Current palm-oil biodiesel output (16,259 tons per year) is sufficient to supply the Guayaquil BRT fleet without requiring additional land. Discussion Although the local biodiesel supply can support the Guayaquil BRT system, extending this bioenergy BRT strategy nationwide would likely generate significant upstream environmental pressures. The findings also highlight variability across emission estimation methods. Conclusion The study shows that while the Guayaquil BRT system can be supplied with existing biodiesel production and evaluated through established emission methodologies, scaling such strategies requires careful consideration of upstream environmental impacts and methodological uncertainty.
