Citi Bike Research Articles

Towards a System Dynamics Framework for Human–Machine Learning Decisions: A Case Study of New York Citi Bike

The growing number of algorithmic decision-making environments, which blend machine and bounded human rationality, strengthen the need for a holistic performance assessment of such systems. Indeed, this combination amplifies the risk of local rationality, necessitating a robust evaluation framework. We propose a novel simulation-based model to quantify algorithmic interventions within organisational contexts, combining causal modelling and data science algorithms. To test our framework’s viability, we present a case study based on a bike-share system focusing on inventory balancing through crowdsourced user actions. Utilising New York’s Citi Bike service data, we highlight the frequent misalignment between incentives and their necessity. Our model examines the interaction dynamics between user and service provider rule-driven responses and algorithms predicting flow rates. This examination demonstrates why understanding these dynamics is essential for devising effective incentive policies. The study showcases how sophisticated machine learning models, with the ability to forecast underlying market demands unconstrained by historical supply issues, can cause imbalances that induce user behaviour, potentially spoiling plans without timely interventions. Our approach allows problems to surface during the design phase, potentially avoiding costly deployment errors in the joint performance of human and AI decision-makers.

A Data-Driven Optimization Framework for Static Rebalancing Operations in Bike Sharing Systems

Bike sharing systems have been widely deployed in urban cities for first- and last-mile transportation. However, because of the geographical and temporal imbalance of bike demand, bikes need to be reallocated system-wide among stations during the night to maintain a high service level while minimizing demand loss due to stockout or overcapacity. Two technical challenges remain in optimizing the static bike rebalancing operations. One challenge is to accurately predict bike pickup and dropoff demand at each station, considering demand substitution effects and subsequently determining the optimal rebalancing quantity for each station. The other is to efficiently optimize the routing of multiple rebalancing vehicles for large-scale bike sharing systems, considering outlier stations with rebalancing quantities exceeding vehicle capacity. To this end, we propose an end-to-end solution to tackle the aforesaid challenges. Specifically, we first develop deep learning-based predictors that capture the time dependencies of station-level demand, the impact of weather conditions, and the demand substitution effect by nearby stations. Based on the demand rate, a sequential simulation-based demand loss estimator is developed to find the optimal rebalancing quantities that lead to the minimum expected demand loss. Then, a mixed integer linear programming model is formulated to optimize the routing problem of rebalancing vehicles. To address the computational challenge, we propose a data-driven decomposition algorithm to support a multivehicle multivisit rebalancing strategy by decomposing the multivehicle routing problem into smaller and tractable single-vehicle routing problems, which can be solved in parallel. Finally, extensive numerical experiments using real-world data from New York City Citi Bike demonstrate the accuracy of the proposed bike demand predictors, the impact of demand substitution, and the efficiency of the data-driven optimization framework. History: Accepted by Ram Ramesh, Area Editor for Data Science & Machine Learning. Funding: This work was supported by the National Natural Science Foundation of China [Grant 72201222] and the Hong Kong Research Grants Council [Grants CityU 21500220 and CityU 11504322]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2022.0182 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2022.0182 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

Financial determinants of mode substitution in residents’ travel behaviour: a case study of public bike-sharing in Lodz, Poland

Bike-sharing networks have achieved considerable success in many cities worldwide, gaining a growing number of supporters for this mode of transportation. While the existing literature covers various aspects related to bike-sharing, the exploration of the relationship between these networks and the financial benefits for residents when transitioning to cycling has been somewhat limited. Therefore, the main objective of our article is to identify the factors influencing urban residents’ decisions to switch to cycling and to understand the significance of financial considerations in shaping changes in travel behaviour.We assessed the perceived affordability of bike-sharing services by measuring respondent satisfaction (via Computer Assisted Personal Interviews – CAPI) with the rental prices of city bicycles. To examine the relationships between variables, we employed statistical tests, including the Fisher test, the chi-square test of independence, and the Mann-Whitney test. Our research findings confirmed that replacing public transportation with bicycles has the most substantial impact, while substituting car trips has a relatively minor effect. Furthermore, our analysis revealed statistically significant associations between price satisfaction and the decision to abandon car travel in favour of cycling, as well as the motivation to save costs and substituting walking and public transport with bicycle travel.

Social Aspects of Smart Urban Mobility

There is a growing interest in sustainable urban transport solutions in cities around the world. These changes, known as “smart urban mobility”, aim to reduce the negative effects of transport on the natural environment and enhance the standard of living for urban dwellers. It should be noted that in addition to this transformation’s technological aspects, modifying the city’s structure and architecture also has a social dimension. The transformation of urban mobility has a significant impact on social relations. Introducing new modes of transport, such as city bikes and electric scooters, creates new ways of moving around the city and can impact social interactions. This can increase social integration and a community’s sense of belonging. At the same time, it is necessary to consider the differences between social groups to ensure equal access to the benefits of these changes in addition to deeper social elements, such as changing the habits and expectations of residents and adapting solutions to the specific needs of each city. This paper aims to look at the social aspects of smart urban mobility, including the impact of these changes on the lives and relationships of city residents.

Powering bikeshare in New York City: does the usage of e-bikes differ from regular bikes?

ABSTRACT In this study, we investigate the difference between shared electric bicycles (e-bikes) and conventional shared bikes operated by Citi Bike in New York City. We examine differences in usage by examining summary statistics and we develop conditional autoregressive models to examine differences in factors associated with trip generation. These factors include bike infrastructure, subway proximity, area demographics, land use, and elevation. We also control for border effects by identifying peripheral stations. The percentage of e-bike trips fluctuates across hours of the day, with a peak around 3 PM but diminishes during peak commuting hours. We find mostly no significant differences in factors associated with the generation of trips between the two modes. Exceptions include a larger effect on e-bike trip generation when the daytime worker population is larger; other differences are due to the racial composition around the station area, some land use categories, and elevation differentials.

Measurement of the lateral characteristics and identification of the Magic Formula parameters of city and cargo bicycle tyres

The paper presents measurements of the lateral force and self-aligning torque from cargo and city bicycle tyres. Based on the experimental data, we have determined the parameters for the Magic Formula model, for lateral force and self-aligning torque. We performed tests with VeTyT, an indoor test rig specific for bicycle tyres, under different vertical loads (ranging from 343 N to 526 N), camber angles (−5, 0, 5) deg and inflation pressure (from 300 kPa to 500 kPa). For each condition, we evaluated the cornering stiffness and found that it generally decreases with the increase in inflation pressure for the tour/city bicycle tyres. However, the cargo tyre tested showed an opposite trend, with an increase in the cornering stiffness as the inflation pressure increased from 300 kPa to 400 kPa.

Sensors capabilities as a creativity tool for engineering product design

Smart and connected physical products enable numerous novel value creations based on embedded sensor data. However, during the product design process, it appears that these value-creation opportunities are not sufficiently explored. To address this gap, a general and accessible ideation tool is proposed to structure a systematic exploration. The tool has been developed from the data sheets of a large number of various embedded sensors. A non-exhaustive list of 14 main sensor capabilities was thus structured to form the core of the tool. During the creative phases of product design, these capabilities are combined with a convenient set of inputs to support the generation of value-creation concepts. A focus group to evaluate the tool was carried out with a medium-sized manufacturer of connected shared city bikes. Expert product designers generated concepts with and without the tool. The results of the experiment suggest that the tool enables the generation of concepts that are more novel and of better quality in comparison to traditional ideation approaches. In addition, participants reported a high level of appreciation for the tool. They also noted a high level of creative exploration and collaboration facilitated by the tool.

Understanding bike-sharing usage patterns of members and casual users: A case study in New York City

Shared bicycle travel has become an important travel mode for urban residents, and bike-sharing platforms are also booming in major cities worldwide. The bike-sharing platform provides users with systematic services: members refer to annual bike-sharing service subscribers and casual users refer to holders of a day pass or single ride ticket. Even though casual users account for a large share of ridership and revenue at bike-share systems in New York City (NYC), very little is known about the characteristics and preferences of casual users and how they compare to members. Based on the open-docked bike-sharing dataset from Citi Bike, we analyze the bike usage patterns of members and casual users in NYC, and how these patterns change in the face of the COVID-19 pandemic on a typical day level. We find that the COVID-19 pandemic has negatively influenced members' bike trip counts on weekdays; bike travel time increases for members during the pandemic and decreases for casual users after the pandemic. To make a profound study concerning spatial heterogeneities, we employ Gaussian Mixture Model (GMM) to cluster the spatiotemporal changes of the station-level bike usage and obtain four clusters for each user type. Combined with the Points of Interest (POI) information, we find that member-related cluster with commuting POIs is significantly affected by the pandemic, while leisure trips are the most severely affected for casual users. Compared with the central area, peripheral clusters with residential and religious POIs are less affected by the pandemic. According to our findings, new operational strategies such as flexible subscriptions can be developed to attract more users, maintain their stickiness, and improve the bike-sharing level of services.

Identifying the health benefits of city bike infrastructure expansion policy

ABSTRACT The aim of this study was to understand how policy actions implemented by city governments to improve infrastructure within the built environment can in turn influence healthy behaviors and population health. The authors investigated the associations between policy actions implemented by city governments to improve biking infrastructure and multiple health outcomes including leisure-time physical inactivity, obesity and coronary heart disease (CHD). In addition, this study tested if bike use for commuting and leisure-time physical activity were mediators between policy actions to improve biking infrastructure and health outcomes, and whether the health effects of bike infrastructure were different in cities of different population density. We merged local sustainability survey data with city-level chronic disease measures for the largest cities in the U.S. (n = 457). Results from regressions using Ordinary Least Squares estimation and mediation tests showed policies implemented to improve biking infrastructure were associated with lower prevalence of leisure-time physical inactivity, obesity and CHD. These relationships were most notable in medium-density cities (1,500–3,000/sq. mi). Moreover, leisure-time physical activity and active commuting mediated the relationship between bike infrastructure improvement policies and health outcomes. This study demonstrated the value of a policy-based intervention in enhancing population health.

FF-STGCN: A usage pattern similarity based dual-network for bike-sharing demand prediction.

Accurate bike-sharing demand prediction is crucial for bike allocation rebalancing and station planning. In bike-sharing systems, the bike borrowing and returning behavior exhibit strong spatio-temporal characteristics. Meanwhile, the bike-sharing demand is affected by the arbitrariness of user behavior, which makes the distribution of bikes unbalanced. These bring great challenges to bike-sharing demand prediction. In this study, a usage pattern similarity-based dual-network for bike-sharing demand prediction, called FF-STGCN, is proposed. Inter-station flow features and similar usage pattern features are fully considered. The model includes three modules: multi-scale spatio-temporal feature fusion module, bike usage pattern similarity learning module, and bike-sharing demand prediction module. In particular, we design a multi-scale spatio-temporal feature fusion module to address limitations in multi-scale spatio-temporal accuracy. Then, a bike usage pattern similarity learning module is constructed to capture the underlying correlated features among stations. Finally, we employ a dual network structure to integrate inter-station flow features and similar usage pattern features in the bike-sharing demand prediction module to realize the final prediction. Experiments on the Citi Bike dataset have demonstrated the effectiveness of our proposed model. The ablation experiments further confirm the indispensability of each module in the proposed model.

Deciphering urban bike-sharing patterns: An in-depth analysis of natural environment and visual quality in New York's Citi bike system

Bike-sharing offers a convenient and sustainable mode of transportation. Numerous studies have investigated the influence of temporal variations in the natural environment on cycling, as well as the impact of physical street characteristics like networks and infrastructures. However, few studies integrated and compared the effects of natural environment and street visual quality on cycling in the spatial dimension. As a case study, we focused on the impact of these two factors on Citi Bike system on weekdays and weekends in New York City, while accounting for sociodemographic and functional factors. This study employed machine learning and multiscale geographically weighted regression models at both station and neighborhood scales for a comprehensive analysis of their relationships. The results reveal that the natural environment factors, particularly visibility, are more important factors associated with bike-sharing use. Among the visual quality factors, motorized traffic has a negative impact on both weekday and weekend cycling. When considering geographical location, sky openness exhibits an unfavorable influence on weekday cycling in specific areas. By combining natural environment and visual quality factors, our study promotes optimal resource allocation and the development of bike-friendly cities.

Cyclomobility: Perspectives of a moving culture

This article is derived from a research effort associated with the concerns present in many works and practices produced today on the subject of mobility, which highlight the importance that the bicycle has been assuming in society. The debate, when inserted in the appreciation of the culture of cyclomobility, tends to reveal new meanings to the contemporary city and to promote the use of bicycles in Brazilian cities as an expression of a culture in motion. In reproducing the character of descriptive and exploratory research, this Article shares some reflections about the growing impact of cyclomobility in the ways of life of the citizen and in the context of the urban space. And in advancing investigations, the analysis aimed to highlight variations in the theme, with results that problematize and project meanings of the use of bicycles in the urban space, aiming at interpreting current trends. It is expected that the reflections presented here can serve as an incentive to discussions on cyclomobility as a culture in motion, inspiring new research and practices.

Demand Prediction of Shared Bicycles Based on Graph Convolutional Network-Gated Recurrent Unit-Attention Mechanism

Shared bicycles provide a green, environmentally friendly, and healthy mode of transportation that effectively addresses the “final mile” problem in urban travel. However, the uneven distribution of bicycles and the imbalance of user demand can significantly impact user experience and bicycle usage efficiency, which makes it necessary to predict bicycle demand. In this paper, we propose a novel shared-bicycle demand prediction method based on station clustering. First, to address the challenge of capturing patterns in station-level bicycle demand, which exhibits significant fluctuations, we employ a clustering method that combines graph information from the bicycle transfer graph and potential energy. This method aggregates closely related stations into corresponding prediction regions. Second, we use the GCN-CRU-AM (Graph Convolutional Network-Gated Recurrent Unit-Attention Mechanism) model to predict bicycle demand in each region. This model extracts the spatial information and correlation between regions, integrates time feature data and local weather data, and assigns weights to the input features. Finally, experimental results based on the data from Citi Bike System in New York City demonstrate that the proposed model achieves a more accurate demand prediction.

Spatiotemporal analysis of bike-share demand using DTW-based clustering and predictive analytics

This paper investigates bike-share activities and explores their relationships with neighborhood features, advancing our current knowledge for integrating cycle facilities into urban space to support first/last-mile mobility. To identify distinct demand patterns, bike stations are clustered based on time-series ridership. Measuring the similarity between time-series data in the transportation field should take into account the influence of phase difference because similar demands happening in the morning and in the afternoon should be considered dissimilar. This study uses a weighted Dynamic Time Warping to address this issue by assigning a larger weight to data points with a larger time difference, enabling a more realistic measure to compare ridership. Using bike-share trip data from Citi Bike, we identified eight station clusters, each of which exhibits unique temporal activities. We further integrated the locations of various points of interest to explore the profiles of the clusters. Bike stations that are closer to public transport, commercial buildings and food service establishments are generally more popular, suggesting that incorporating spatial contexts can develop a richer understanding of bike-share usage. In addition, this study goes beyond descriptive analytics by investigating the role of neighborhood features in predicting cluster memberships of bike stations. Our results show that non-tree-based models, such as Support Vector Machines and K-Nearest Neighbors, outperform tree-based models. This study provides valuable insights for both urban planners and bike-share operators. When assessing the potential of new bike-share infrastructure, urban planners can deploy our models to identify the cluster-specific demand pattern based on neighborhood features. Bike-share operators can also utilize our findings to identify neighborhoods that require strategic supply of bikes and parking space that may vary within a day.

Learning from Imbalanced Datasets: The Bike-Sharing Inventory Problem Using Sparse Information

In bike-sharing systems, the inventory level is defined as the daily number of bicycles required to optimally meet the demand. Estimating these values is a major challenge for bike-sharing operators, as biased inventory levels lead to a reduced quality of service at best and a loss of customers and system failure at worst. This paper focuses on using machine learning (ML) classifiers, most notably random forest and gradient tree boosting, for estimating the inventory level from available features including historical data. However, while similar approaches adopted in the context of bike sharing assume the data to be well-balanced, this assumption is not met in the case of the inventory problem. Indeed, as the demand for bike sharing is sparse, datasets become biased toward low demand values, and systematic errors emerge. Thus, we propose to include a new iterative resampling procedure in the classification problem to deal with imbalanced datasets. The proposed model, tested on the real-world data of the Citi Bike operator in New York, allows to (i) provide upper-bound and lower-bound values for the bike-sharing inventory problem, accurately predicting both predominant and rare demand values; (ii) capture the main features that characterize the different demand classes; and (iii) work in a day-to-day framework. Finally, successful bike-sharing systems grow rapidly, opening new stations every year. In addition to changes in the mobility demand, an additional problem is that we cannot use historical information to predict inventory levels for new stations. Therefore, we test the capability of our model to predict inventory levels when historical data is not available, with a specific focus on stations that were not available for training.

Assessing the role of public transportation to foster city bike tourism. The case of Italy

Sustainability issues promote the combination of greener transport means when visiting urban environments. This paper focuses on bike tourists by analysing their choice of cities as a destination and the decision to use public transportation to connect within-destination places during holidays. By using data from an Italian survey on bike tourism in 2020, a bivariate probit model is used to study the role of socio-demographics, bike-related factors, travel characteristics, and cycling and accommodation features. The odds of visiting cities are positively affected by the length of stays, available commercial and bike recovery services, while the negative effect of road traffic is confirmed. Instead, combining bikes with public transportation is more likely for low-cost tourists, mainly lodging in B&Bs, and for people with higher sensitivity to bike-related services. From a transport policy perspective, our results support claims for sustainability measures jointly affecting tourists’ decisions on destinations and travel mode choices.

Encouraging sustainable mobility: community case study on workplace initiatives in Lahti, Finland

The transport sector represents about 30 per cent of all carbon dioxide (CO2) emissions in developed countries. Developing sustainable transport and mobility play a central role in reducing greenhouse gas emissions. Sustainable mobility means a mind shift where transport in private cars is replaced by different modes of more sustainable mobility, such as, walking, biking, and public transport. The transformation toward more sustainable mobility plays a key role in reaching CO2 emission reduction goals. However, in addition to the environmental perspective, also social and economic aspects are interconnected in the change. In this sustainability shift, employers can encourage the employees through offering and supporting new alternatives for mobility. This article aims to study how different sustainable mobility initiatives provided by an employer are adopted by employees. This case study presents a set of pilots implemented at a workplace in Lahti, Finland in May-October 2022. The mobility forms offered for commuting were fringe benefits from employment, that is, employer-subsidized commuter tickets and employer-provided bicycle benefits. Travel during the workday was supported through introducing the use of shared electric city bikes and scooters. The research data consisted of short surveys before (n = 70) and at the end of the pilots (n = 66), and thematic interviews (n = 8) during the implementation period. The pilots were implemented in collaboration with the employer and two local universities. Before the pilots, the main part of the employee participants was using private cars for commuting and travel during the workday. Results show that the pilots were successful in introducing more sustainable ways of mobility. However, the employer's role in preparing, supporting, and planning the continuation of support for sustainable commuting and mobility is essential.

Resources Relocation Support Strategy Based on a Modified Genetic Algorithm for Bike-Sharing Systems

In recent decades, special attention has been given to the adverse effects of traffic congestion. Bike-sharing systems, as a part of the broader category of shared transportation systems, are seen as viable solutions to these problems. Even if the quality of service in bike-sharing service systems were permanently improved, there would still be some issues that needed new and more efficient solutions. One of these refers to the rebalancing operations that follow the bike depletion phenomenon that affects most stations during shorter or longer time periods. Current work develops a two-step method to perform effective rebalancing operations in bike-sharing. The core elements of the method are a fuzzy logic-controlled genetic algorithm for bike station prioritization and an inference mechanism aiming to do the assignment between the stations and trucks. The solution was tested on traffic data collected from the Citi Bike New York bike-sharing system. The proposed method shows overall superior performance compared to other algorithms that are specific to capacitated vehicle routing problems: standard genetic algorithm, ant colony optimization, Tabu search algorithm, and improved performance compared to Harris Hawks optimization for some scenarios. Since the algorithm is independent of past traffic measurements, it applies to any other potential bike-sharing system.

Equity Analysis of Bikeshare Access: A Case Study of New York City

This study investigates the association between bikeshare program usage and neighborhoods with different socio-demographic information. This study investigated 2023 Citi Bike bikeshare data in New York City. For the analysis, Generalized Linear Regression (GLR) in ArcGIS Pro was used. The results revealed that Citi Bike bikeshare ridership increases in neighborhoods with higher income, more workers, more males, higher density, and lower African American and Hispanic population in New York City.

Micromobility Vehicles, Obstructions, and Rider Safety Behaviors in New York City Bike Lanes.

Micromobility vehicles (MMVs) have become increasingly popular, particularly in urban areas where infrastructure has improved in recent years to facilitate their use. The purpose of this study was to observe protected bike lanes in 10 zones of Manhattan, NYC to: (1) describe the MMVs in bike lanes by type, phone and helmet use; and (2) document MMV users' responses to obstructions. Approximately 1 in 4 of all riders (260/998) were wearing a helmet. Fewer than 2% were observed using a phone while moving. Fewer than 9% of Citi Bike users were wearing a helmet. In contrast, over one-third of non-Citi Bike users were wearing a helmet (228 of 670, 34.03%). This difference was determined to be significant by a chi-squared test (a = 0.05) with a p-value less than 0.0001. Of the 988 MMVs observed in this study, 398 (40.28%) were motorized and 590 (59.72%) were non-motorized. A similar proportion of users of motorized riders versus non-motorized vehicles were wearing a helmet (28.14%, 112/398 versus 24.41%, 144/590). A total of 232 riders (23.50%) encountered an obstruction in their bike lane. Of these obstructions in a bike lane, 82.33% (191/232) were a car/vehicle and 17.67% (41/232) was garbage. A large majority of riders (87.93%) reacted by riding into the traffic lane. These findings suggest that further research and local education, enforcement, and legislative efforts are needed to examine and implement best practices in the safe operation of MMVs, decreasing bike lane obstructions, promoting helmet use, and raising awareness of MMV legislation.