Travel Behaviour and Society最新文献

Shared Autonomous Vehicle (SAV) has many impacts on the transport development, such as saving parking space. However, SAV meets a huge challenge in terms of vehicle supply and user demand imbalance. The traditional mathematical optimization method cannot be well used due to the computational burden. Hence, this paper proposes a Reinforcement Learning (RL) based SAV relocation approach. First, two types of RL agents, car-based and zone-based agents, are developed as agents for vehicles and stations, respectively. Then, the RL scheme is trained by using historical demand data to facilitate real-time carsharing relocation. Finally, to compare the proposed two types of RL methods, three scenarios are used: small-scale, middle-scale, and large-scale networks. Solutions indicate that the enhanced zone-based method achieves an additional 146% profit compared to traditional threshold-based relocation strategies. The user travel behaviour impacts are provided by analyzing parking demand and travel movements among residential, industrial and commercial zones.

This paper seeks to fill in the current evidence gap on the relationship between travel satisfaction and e-scooter use. To do so, online survey data from e-scooter users of a University community in the city of Thessaloniki, Greece were collected and statistically analyzed. In line with previous research on travel satisfaction, the results of a series of exploratory and confirmatory factor analyses unveiled the potential of factors surrounding positive activation, positive deactivation, and cognitive evaluation to serve as key determinants of satisfaction of e-scooter riders. Furthermore, a structural equation model (SEM) was estimated to identify the impact of user- and trip-related characteristics on travel satisfaction. The results of the SEM showed that older and high-income riders, those with good self-reported physical condition, riders of private e-scooters and leisure travelers are more likely to perceive greater satisfaction by the use of their e-scooters. The findings of this study overall contribute to the current state-of-knowledge relating to travel satisfaction and pave the way for potential policy actions that could improve users’ experience with e-scooters and enhance the overall attractiveness of micromobility in the pathway towards sustainable and inclusive mobility.

Electric bicycles (e-bikes) are gaining popularity globally as green and active modes of transport. Research on e-bike adoption to date has predominantly investigated various extrinsic motivations to use an e-bike, and little is known about the psychological influences. Examining the influence of psychological determinants is crucial to comprehend the unobserved individual and social factors affecting e-bike adoption. This study aims to determine the sociopsychological influences associated with e-bike use among the daily commuters (employees) of two university campuses in India which represent self-reliant societies. The study extends the classic Theory of Planned Behaviour (TPB) framework to account for influences like personal norms, knowledge of e-bikes, and parking anxiety, in addition to fundamental TPB constructs (attitudes, subjective norms, and perceived behavioural control). Results are constructed by a structural equation model (SEM) for a sample of 347 non-users of e-bikes. SEM model indicates that among all TPB constructs, subjective norms are the strongest predictors of intentions to use e-bikes. Both subjective norms and attitudes have a significant positive effect on adoption intentions, whereas perceived behavioural control has a significant negative effect. Personal norms had significant indirect positive effects on e-bike use intentions with fundamental TPB constructs as mediators. Interestingly, parking anxiety (anxiety from missing the car parking spaces) positively, and knowledge of e-bikes negatively influence the e-bike adoption intentions. Further, comparison by psychological influences reveals that a significant difference exists between the perceptions of the two university commuters towards e-bike use, possibly due to the distinct terrain and climate characteristics of the universities. The study findings have implications for designing effective e-bike interventions and awareness programs seeking to promote e-bike adoption, particularly in self-reliant communities.

Distracted driving substantially undermines road safety, and the frequent, low-accuracy distracted driving risk prediction alarms may detrimentally affect drivers’ judgments. The study aims to improve risk prediction accuracy by mapping the distinctive interactions of eye glances and manual sequences in distracted driving into a graph structure of nodes and edges. Distraction patterns are categorized as visual distractions (VD), manual distractions (MD), and visual-manual distractions (VMD) based on daily driving behaviours, such as eating, drinking, reaching for something, and using mobile phones. 1,806 distraction alarm records came from an active safety platform (ASP) in Beijing, covering 69 drivers from 23 hazardous materials road transport companies. The study extracts characteristics to assess distracted driving risks, including visual, manual, and driving performance features. Subsequently, unsupervised learning is used to cluster risk features into three categories (low, medium, and high), which serve as labels for the risk prediction model. In addition, each distraction alarm sequence is divided into nodes and edges of a graph. More specifically, five visual areas, forward (F), object of distraction (Dis), left window (LW), rear-view mirror (RM), center dashboard (C), as well as manual sequences, single hand (H), double hands (2H) are represented as nodes. The edges are connected in parallel (occurring simultaneously) or in series (occurring sequentially), with arrows pointing from the earlier node to the later one. Furthermore, coupled with time, global, and environmental features, a temporal graph attention network (TGAT) with integrated time functions and multi-head attention mechanisms is developed to predict distracted driving risks. The results indicated that VMD demanded more visual and manual resources and led to more high-risk alarms than VD and MD. Besides, TGAT reached a promising result, outperforming other time series methods. This study is valuable for driver distraction monitoring and driving risk assessment, significantly contributing to the enhancement of road safety.

This study examines the charging location choice behavior of residents in relation to employer-provided charging facilities and investigates the impact of different government subsidy strategies for charging facility construction on urban spatial structure and traffic-related air pollution. Using a monocentric two-zone city model, we analyze the residential distribution and travel mode choice of urban residents under various subsidy strategies from the perspective of social planners. The government can optimize urban resource allocation through transport infrastructure investments and implementation of emission taxes to maximize the so-called residents’ utility. We observe that despite the distorting effect of employer-provided charging facilities on travel mode choice, residents still experience an enhanced utility due to the positive environmental impact resulting from reduced pollution levels associated with electric vehicles (EVs) compared to gasoline vehicles (GVs). We provide numerical evidence to show that government subsidies, whether directed towards companies or residents, can effectively enhance the adoption of EVs and improve utility. In low-density cities, it is advisable for the government to encourage and subsidize residents in building charging facilities. Conversely, in high-density cities, prioritizing subsidies for companies would be more beneficial in improving utility and expanding the coverage of charging facilities. Furthermore, in cities where resident environmental awareness is relatively low, it is recommended that the government increase subsidies to promote EV adoption.

Previous research has found that components of the natural and built environment play an important role in regulating ambient temperature. However, existing research regarding the association between these environmental characteristics and thermal exposure has focused mainly at the macro level, leaving this relationship at the individual level underexplored. It remains unknown how the environment functions differently in determining thermal exposure among various types of trips and how this mechanism differs by gender. Using GPS walking trajectory data collected in Nanjing, China, this study examines the extent to which male and female pedestrians experience different levels of thermal exposure, and how the thermal exposure determinants work differently between utilitarian and recreational trips. Descriptive analysis shows that men experience higher per-minute thermal exposure than women, and both male and female pedestrians face higher thermal exposure per minute during utilitarian walks compared to recreational walks. Generalized linear mixed model results indicate that green spaces significantly reduce thermal exposure for both male and female pedestrians during utilitarian walking trips, but this effect only works among women regarding recreational walking. We also identified a negative relationship between water bodies and thermal exposure during recreational walks, but this correlation only occurs among women. Our study suggests that the natural environment’s mitigating effect on thermal exposure differs by gender among different types of walking trips. Policymakers should consider these disparities to avoid exacerbating gender inequality in the arena of thermal exposure and health.

Due to the issues of land redevelopment and changes of use within urban areas, many cities must adopt measures to reorganise and optimise parking space. This paper proposes a methodology to study one of them by implementing parking information systems (PIS). This solution offers users a competitive advantage by allowing them to know about the free parking spaces at the moment of decision-making. To achieve this goal, microscopic simulations are conducted to analyse the effects of various scenarios involving the implementation of PIS. The data used in these simulations is obtained from the Santander area in Spain. For the evaluation of results, a methodology has been developed that combines the evaluation of social factors for citizens and operational impacts for decision-makers. The results show significant improvements with increasing user information rate, e.g., the number of unsuccessful parking attempts before finding a final parking space is reduced by 55%, and 37% less particulate pollutants are emitted into the atmosphere.

Most ride-sourcing platforms, exemplified by industry leaders like Uber, Lyft, and Didi, provide a range of ride services tailored to the diverse preferences of their passengers. Passengers, driven by their distinct priorities, may opt for high-class (HC) ride services, such as Luxury rides, if they value service quality, while those more cost-conscious may gravitate toward low-class (LC) ride services, including basic solo and shared rides. However, this market fragmentation can manifest as an excess of HC vehicles idly cruising the streets, while an insufficient number of LC vehicles struggle to meet passenger demand for LC services. To mitigate this issue, upgrading strategy is proposed where some LC vehicle requests are elevated to HC ride services without incurring additional charges. This study embarks on an initial exploration of the impacts of upgrading within the ride-sourcing system. We develop a mathematical model to depict the equilibrium conditions and analyze the collective influence of operational strategies, encompassing upgrading, spatial pricing, and vehicle repositioning, on system performances. Our research identifies scenarios in which the platform should employ these strategies to balance supply and demand and curb superfluous idle vehicle movements, supported by both theoretical and numerical analyses. The results offer operational insights that guide platform decisions, allowing them to adapt their strategies effectively in response to various supply–demand dynamics.

Recently, several new concepts and innovative technologies have emerged to overcome the problems of urbanization, which can be hardly solved with using exclusively private vehicles or conventional public transport services. One of the new solutions is the Mobility-as-a-Service (MaaS) concept, a user-centric mobility distribution scheme, in which the user needs are satisfied via a single platform, and multiple transport options are offered by one MaaS operator (MO). In the last years, a couple of MaaS pilots were performed, but previous papers fail to focus on the pilot development and the proper description of the minimum viable product. A pilot of MaaS in Budapest has been developed by using the innovative Scrum methodology successfully involving six mobility service providers, such as public transport, shared mobility, and taxi, in the live demonstration. Current article provides detailed information about the pilot development including technical, legal, and business use cases for all service providers. The results of the recruitment and the characterization of the early-bird users are presented, too. The iterative pilot development process can be utilized by those MOs and governmental organizations that would like to initiate a new mobility project based on the MaaS concept.

In response to the absence of demographics in increasingly emerging big data sets, we propose a novel method for inferring the missing demographic information based on similarity in people’s daily multi-dimensional activity-travel patterns as well as the characteristics of the area they move about. Instead of using isolated activity-travel attributes to infer social demographic features, our proposed method first calculates the similarity of people’s multidimensional daily activities and travels as well as characteristics of their visiting locations, between those for whom the social demographics are to be imputed (target) and those with known demographics (base) using a polynomial function. The weights of the function are determined using the permutation feature importance method, and then dynamic time warping is used to align the multidimensional activity sequences of the base and target sample and measure their similarities. For each person in the target database, a matched list is created consisting of those with the most similar activity-travel sequences in the base sample. A support vector machine is then trained using the base sample as input to impute the demographics of the target sample. The proposed model is trained using a national travel survey and validated by applying it to a GPS dataset. The results show that the proposed method outperforms existing methods in predicting four selected demographics: gender, age, education level, and work status, with an accuracy range between 91% and 94% for the national dataset and 88% to 91% for the GPS data. This study highlights the importance of considering the multidimensional and sequential nature of peoples’ daily activity-travel patterns in the imputation of demographic features.