Search

Taxis spend a large amount of time idle, searching for passengers. The routes vacant taxis should follow in order to minimize their idle times are hard to calculate; they depend on complex effects like passenger demand, traffic conditions, and inter-taxi competition. Here we explore if reinforcement learning (RL) can be used for this purpose. Using real-world data from three major cities, we show RL-taxis can indeed learn to minimize their idle times in different environments. In particular, a single RL-taxi competing with a population of regular taxis learns to out-perform its rivals.

Bicycling has grown significantly in the past ten years. In some regions, the implementation of large-scale bike-sharing systems and improved cycling infrastructure are two of the factors enabling this growth. An increase in non-motorized modes of transportation makes our cities more human, decreases pollution, traffic, and improves quality of life. In many cities around the world, urban planners and policymakers are looking at cycling as a sustainable way of improving urban mobility. Although bike-sharing systems generate abundant data about their users’ travel habits, most cities still rely on traditional tools and methods for planning and policy-making. Recent technological advances enable the collection and analysis of large amounts of data about urban mobility, which can serve as a solid basis for evidence-based policy-making. In this paper, we introduce a novel analytical method that can be used to process millions of bike-sharing trips and analyze bike-sharing mobility, abstracting relevant mobility flows across specific urban areas. Backed by a visualization platform, this method provides a comprehensive set of analytical tools to support public authorities in making data-driven policy and planning decisions. This paper illustrates the use of the method with a case study of the Greater Boston bike-sharing system and, as a result, presents new findings about that particular system. Finally, an assessment with expert users showed that this method and tool were considered very useful, relatively easy to use and that they intend to adopt the tool in the near future.

As is often believed that the more centrally located a shop, the higher its sales volume, this paper analyzed relationships between the spatial clustering of retail stores, their respective transaction volumes, and the urban street networks to determine whether, and to what extent, the accessibility and density of a store’s location was correlated with its transaction volume. While this hypothesis is widely accepted, its veracity is underexplored and rarely validated using large-scale empirical datasets, possibly owing to the lack of access. Therefore, transaction datasets and accessibility indicators were first examined; a clear, positive correlation between density and revenue was found for specialty stores wherein people do “comparison shopping,” and for stores that complemented each other for activities such as “one-trip shopping,” the revenues were positively correlated when the stores were clustered. Generally, daily-use stores’ revenues were more sensitive to local access and those of non-daily-use stores were more sensitive to global access. In conclusion, these findings would not have been found using conventional methodology focused on the retail sector as a whole, because aggregate market mechanisms would have hidden the observed effects on specific store categories. Therefore, upon disaggregating the data, we found a distinct heterogeneity across the different store types for what concerns the relationship between revenue and location.

The development of mobile Internet, smartphones, and location-based services has enabled ridesourcing, which pools vehicles and drivers to provide on-demand travel services. As an alternative transportation option, ridesourcing has significant impacts on urban travel. However, the unique mobility pattern of ridesourcing and its impact on vehicle electrification have not been well studied. To address this gap, this paper presents a comparative, big-data-driven framework to characterize the ridesourcing mobility pattern, and evaluate the acceptance potential of electric vehicles for ridesourcing in comparison with other types of vehicle use. Multi-temporal resolution ridesourcing trips are extracted from raw GPS trajectories. The patterns of three urban travel (household, ridesourcing, and taxis) are extracted from GPS trajectories in Beijing, and compared. The electrification potentials of these types of travel under different charging levels are then evaluated. The results demonstrate that mobility patterns of household, ridesourcing, and taxi drivers are similar when a single trip is considered but differ significantly when total vehicle travel is considered. We show that potential acceptance of electric vehicles decreases significantly from household to ridesourcing and taxi vehicle use. These findings provide useful insights into of the role vehicle electrification can play in sustainability of urban personal transportation across a range of drivers.

As a clean and renewable resource, solar energy is increasingly being used to relieve the pressures on environmental protection and the exhaustion of conventional energy. Although photovoltaic modules have been installed in many cities, the lack of quantitative mapping of the annual solar energy potential of urban surfaces hinders the effective utilization of solar energy. Herein, we provide a solar irradiation estimation solution for three-dimensional (3D) cities to quantify annual irradiations on urban envelopes and to investigate the effect of urban morphology on the resulting solar capacity. By modelling urban surfaces as 3D point clouds, annual irradiations of the point clouds were estimated. An empirical investigation across ten cities suggests that urban areas at lower latitudes tend to have larger values of annual irradiation; moreover, an area having greater building heights consistently has the largest third quartile of irradiation compared with lower buildings in the same city. Conversely, areas with many low buildings have a larger proportion of useable areas; in this arrangement, facades can optimally utilize solar energy, meaning that large irradiations are concentrated on certain facades. The Pearson correlation coefficients between solar capacity and urban morphology indices suggest that urban morphology has an important effect on solar capacity.

Throughout history, social movements have often been catalysts for radical societal change. In the past two decades, hashtag activism, the use of social media platforms for internet activism, has become a driving force behind the development of social movements across the world. In the field of social movements science, a large body of research has studied the role of hashtag activism for the formation of social movements, but less efforts have been allocated towards the study of the spatio-temporal relationship that exists between hashtag activism and political mobilization processes. In this study, a large geo-located social media dataset pertaining to the #JusticeForGeorgeFloyd social movement is used to quantify the spatio-temporal relationship between hashtag activism and physical protest activity. Results of this study indicate that at the national, state, and county scale, hashtag activism bears a strong positive temporal relationship with physical protest activity. Through the statistical modeling of this relationship, it is possible to predict the intensity of national, state, and county level protest activity on a given day based on the intensity of state and county level social movement related tweet counts the day before. At smaller scales (i.e. county), however, this research finds that the strength of this relationship is more variable, and it appears to be strongly dependent on population density. As such, this study reveals the existence of a critical population density threshold of 200 people/km2 , at which an increase in the number of social movement related tweets is likely to spark with an increase in the number of physical protests. These results represent novel insights that further contribute to our understanding of the complex spatio-temporal political mobilization processes that occur during social movements.