Taxi Trips Research Articles

Multiple Information Spatial–Temporal Attention based Graph Convolution Network for traffic prediction

Traffic prediction (forecasting) is a key problem in intelligent transportation. It helps engineers to obtain traffic trends in advance so that they can make favorable decisions quickly and accurately, or even improve the working mode of existing systems. However, it is very challenging to design a model for such problem that fully utilize the factors related to traffic. This paper investigates machine learning in traffic prediction and proposes Multiple Information Spatial–Temporal Attention based Graph Convolution Networks (MISTAGCN). The model consists of two parts. The first part utilizes combinations of different inputs and graph structures to compute the corresponding latent variables. In the second part of the model, the latent variables are comprehensively integrated and deeply mined. In particular, as the basic component of the model, a STBlock integrates mechanisms such as temporal attention, spatial attention, graph convolution, ordinary convolution and residual connections to fully explore the potential information contained in the data. Experiments on Haikou online car-hailing dataset and New York yellow taxi trip dataset illustrate that the proposed model outperforms state-of-art baseline models.

Recognizing mixed urban functions from human activities using representation learning methods

ABSTRACT When various urban functions are integrated into one location, they form a mixture of functions. The emerging big data promote an alternative way to identify mixed functions. However, current methods are largely unable to extract deep features in these data, resulting in low accuracy. In this study, we focused on recognizing mixed urban functions from the perspective of human activities, which are essential indicators of functional areas in a city. We proposed a framework to comprehensively extract deep features of human activities in big data, including activity dynamics, mobility interactions, and activity semantics, through representation learning methods. Then, integrating these features, we employed fuzzy clustering to identify the mixture of urban functions. We conducted a case study using taxi flow and social media data in Beijing, China, in which five urban functions and their correlations with land use were recognized. The mixture degree of urban functions in each location was revealed, which had a negative correlation with taxi trip distance. The results confirmed the advantages of our method in understanding mixed urban functions by employing various representation learning methods to comprehensively depict human activities. This study has important implications for urban planners in understanding urban systems and developing better strategies.

Two-Sided Deep Reinforcement Learning for Dynamic Mobility-on-Demand Management with Mixed Autonomy

Autonomous vehicles (AVs) are expected to operate on mobility-on-demand (MoD) platforms because AV technology enables flexible self-relocation and system-optimal coordination. Unlike the existing studies, which focus on MoD with pure AV fleet or conventional vehicles (CVs) fleet, we aim to optimize the real-time fleet management of an MoD system with a mixed autonomy of CVs and AVs. We consider a realistic case that heterogeneous boundedly rational drivers may determine and learn their relocation strategies to improve their own compensation. In contrast, AVs are fully compliant with the platform’s operational decisions. To achieve a high level of service provided by a mixed fleet, we propose that the platform prioritizes human drivers in the matching decisions when on-demand requests arrive and dynamically determines the AV relocation tasks and the optimal commission fee to influence drivers’ behavior. However, it is challenging to make efficient real-time fleet management decisions when spatiotemporal uncertainty in demand and complex interactions among human drivers and operators are anticipated and considered in the operator’s decision making. To tackle the challenges, we develop a two-sided multiagent deep reinforcement learning (DRL) approach in which the operator acts as a supervisor agent on one side and makes centralized decisions on the mixed fleet, and each CV driver acts as an individual agent on the other side and learns to make decentralized decisions noncooperatively. We establish a two-sided multiagent advantage actor-critic algorithm to simultaneously train different agents on the two sides. For the first time, a scalable algorithm is developed here for mixed fleet management. Furthermore, we formulate a two-head policy network to enable the supervisor agent to efficiently make multitask decisions based on one policy network, which greatly reduces the computational time. The two-sided multiagent DRL approach is demonstrated using a case study in New York City using real taxi trip data. Results show that our algorithm can make high-quality decisions quickly and outperform benchmark policies. The efficiency of the two-head policy network is demonstrated by comparing it with the case using two separate policy networks. Our fleet management strategy makes both the platform and the drivers better off, especially in scenarios with high demand volume. History: This paper has been accepted for the Transportation Science Special Issue on Emerging Topics in Transportation Science and Logistics. Funding: This work was supported by the Singapore Ministry of Education Academic Research [Grant MOE2019-T2-2-165] and the Singapore Ministry of Education [Grant R-266-000-135-114].

Zone-level traffic crash analysis with incorporated multi-sourced traffic exposure variables using Bayesian spatial model

The primary objective of this study is to discover traffic exposure variables from some new data sources and explore how these new data sources and their combination affects the performance of zone-level crash models. Seven types of check-in activities and five types of taxi trips are inferred from Twitter and taxi GPS records, respectively. Then, Bayesian spatial models are employed to conduct zone-level traffic crash analysis. The results suggest that some specific check-in activities and inferred taxi trips are closely related with zone-level crash counts, and thereby confirms the benefits of incorporating new data sources into zone-level crash models. The comparative analyses further indicate that twitter check-in activities perform better than inferred taxi trips as a proxy for traffic exposures on spatial analyses of traffic crashes, and detailed trip purpose information hidden in new data sources greatly benefit zone-level crash models than simply aggregating location points in each zone. The results of this study reveal that each big data source has its prominent coverage of user groups and spatial areas, and their combination can serve as effective supplementary information to traditional exposure variables to improve the performance of zone-level crash models and better reveal the spatial impacts of human activities on traffic crashes. The findings of this study can help transportation authority develop more targeted traffic demand adjustment strategies to effectively reduce zone-level crash risks.

A Decision Support System for Taxi Drivers

The taxi sector is being revolutionized worldwide due to innovative companies, such as Uber, Lyft and more, that are trying to acquire a share of the market. In the same context, contemporary businesses in Greece have started to expand their activities. This present research is subsumed into the concept of innovation in the conventional taxi sector in Greece and attempts to implement pioneering ideas, to offer better services for taxi customers and profitable financial circumstances for taxi owners and drivers. The main data source regarding the taxi trips are the Hellenic Institute of Transport and Taxiway, the largest taxi fleet management company that operates in Thessaloniki. The given data consisted of 170 taxi trips originating from a taxi stand in the city center, 1981 routes which began from the area around the taxi stand and 1832 trips of taxis that passed through the examined area near the stand. These data, in accordance with a financial analysis of taxis’ operational costs, have been processed to produce statistical results, such as the mean waiting time in each stand or possible spots for a taxi to pick up a customer. The research relies on these data, for the sake of building an application in which taxi drivers would be advised on their dilemma of choosing between waiting in taxi stands or hailing in the surrounding area to find a customer. With the utilization of such a website, drivers can compare the predicted benefit values for hailing or standing alternatives aiming to optimize the taxi downtime. The analysis concluded that the system can potentially increase the average expected profit per trip, even by 50% in a comparison to a different decision taken by the taxi driver.

Acute illness symptoms among investment professionals and stock market dynamics: Evidence from New York City

In the U.S., stock market professionals (e.g., traders, portfolio managers, and analysts) are clustered in New York City (NYC). In view of this, I exploit daily changes in the incidence of acute illness symptoms among 18–64 year old New Yorkers to identify exogenous variation in the rate of acute illness among market professionals and estimate its causal impact on key stock market outcomes. A detailed analysis of taxi trips from a sample of financial institutions to local hospitals provides support for my identification assumption. Other things equal, increased rates of acute physical illness (i.e., reduced productivity) among market professionals hamper price discovery and lower trading activity, volatility, and returns. A one-standard-deviation increase in my illness incidence proxy reduces by 18% (6.7%) the immediate response of stock prices to earnings surprises (changes in analysts’ consensus recommendations) and increases by 29% (42%) their delayed response.

Neighborhood retail amenities and taxi trip behavior: A natural experiment in Singapore

While a small change in land use planning in existing neighborhoods may significantly reduce private vehicle trips, we do not have a great understanding of the magnitude of the project- and shock-based causal change in travel behaviors, especially for the retail purpose. We analyze the impact of newly developed malls on the retail trip behavior of nearby residents for shopping, dining or services. Using the difference-in-differences approach and big data from a major taxi company in Singapore, we find that households residing within 800 m from a new mall are significantly less likely to take taxis to other retail destinations after the mall's opening. This travel behavior change encompasses both intensive (the share of the number of retail trips out of total taxi trips originating from each residential building) and extensive (the share of retail trip distance out of total taxi trip distance) margin responses. We further demonstrate that the magnitude of this retail trip reduction is more significant during PM peak hours and for residential buildings that are located farther from the Central Business District and in less self-sufficient communities. Our research provides suggestive evidence on the significant role of neighborhood retail amenities to the change in mode and destination choices for retail. An important implication is that improving self-sufficiency for suburban neighborhoods could not only enhance the well-being of their residents but also increase aggregate welfare by reducing the level of congestion. This is extremely relevant to residential town planning in both developing and developed countries.

Effect of Social Vulnerability on Taxi Trip Times during Hurricane Sandy

The increase in the availability of GPS-based movement data has enabled the exploration of mobility patterns in urban transportation networks. Understanding the relationship between social vulnerability and transportation flows from big data during natural disasters is crucial for utilities and policymakers for decision-making purposes, such as evacuation and restoration planning. In this study, we explore the geographic variation of changes in trip times of taxi trips in New York City (NYC) before and after Hurricane Sandy (2012) using GPS trajectory data in relation to the underlying socio-economic distribution of impacted populations using localized regression technique with GWR. The findings reveal how the spatial patterns of trip change times with respect to SVI, income levels and population density in NYC.

Desctiptive Study of Taxis Services in Duhok City

Taxis in the city of Duhok are widely used as an alternative public transportation service due to the minimal or non-existent usage of public transportation by expats and locals, and the predominance of private cars. In this study a questionnaire survey was conducted targeting the study area citizens, with a sample size of 500 taxi passengers. The forms were distributed to the respondents in different areas within the city. From the questionnaire survey it was found that 40% of the respondents had one or two trips during the day by taxi, where the morning period recorded the highest percentage by 63%. This is due to the fact that 65% of these trips are for the purpose of going and returning from work. To improve the taxi service in Duhok city 41% of the respondents encouraged the use of Uber as the best solution because in their opinion Uber will reduce the congestion inside the city and thereby reduces the percentage of pollution. 41% of the respondents do not have a bad experience with a taxi, while 25% of them have one bad experience. Therefore, some citizens specially women don’t use taxis, firstly, because of the customs and traditions and secondly, fear of bad experiences that may happen with them. One of the reasons of delaying the taxi trip was the city congestion which corresponds to the highest percentage of 78%. Finally, the interviewer asked if the respondent used public buses and if they were available in the city. Participants who claimed that they use buses were 58%

Exploring the Spatio-Temporal and Behavioural Variations in Taxi Travel Based on Big Data during the COVID-19 Pandemic: A Case Study of New York City

The COVID-19 pandemic has caused severe social and economic chaos worldwide. To explore the impact of the COVID-19 pandemic on the travel patterns of residents, we analysed taxi trajectory data and COVID-19 pandemic data from New York City. Pearson coefficients, which were −0.7139, −0.8041, and −0.7046 during the three waves of the COVID-19, revealed a significant negative correlation between confirmed cases and taxi trips. Moran’s I was higher in drop-off areas than in pick-up areas, indicating a stronger spatial autocorrelation in drop-off areas during the study period. The hotspots of travel destinations had changed by spatial clustering, and variations in origin–destination distribution were obvious after the pandemic. Comparison of temporal and spatial dimensions before and after the pandemic revealed that strict epidemic policies directly affected travel. For instance, a week after the restrictions the taxi journeys plummeted by 95.3%, and their spatial and temporal patterns also changed. Once the anti-epidemic policy was eased or lifted, the taxi travel recovered, whereas, notably the new Omicron wave did not cause dramatic changes in taxi journeys. Despite this, travel spatial and temporal patterns did not return to pre-pandemic levels by the end of March 2022, the taxi journeys remained below half the pre-pandemic level. This study identified the profound impact of the COVID-19 outbreak on travel patterns and revealed distinct variations in behavioural responses during the pandemic and in response to subsequent policies. Strengthening targeted epidemic prevention and control measures are required to improve the balance between anti-epidemic policies and implementation efforts, that will facilitate the recovery of urban transport, work, and lifestyle of residents.

A Regression Model of the Number of Taxicabs in U.S. Cities

In cities that control the number of taxicabs by law or regulation, setting the number of cabs is one of the most important decisions made by taxicab regulators and elected officials. Licensing either too many or too few cabs can have serious deleterious effects on the availability and quality of service and the economic viability of the taxi business. Yet local officials often have difficulty quantifying the demand for taxi service or tracking changes in demand.Multiple regression modeling of the number of cabs in 118 U.S. cities identifies three primary demand factors: the number of workers commuting by subway, the number of households with no vehicles available, and the number of airport taxi trips. These results can be used to identify peer cities for further comparison and analysis and to guide regulators in measuring changes in local demand for cab service.

Fan-Shaped Model for Generating the Anisotropic Catchment Area of Subway Stations Based on Feeder Taxi Trips

The catchment areas of subway stations have always been considered as a circular shape in previous research. Although some studies show the catchment area may be affected by road conditions, public transportation, land use, and other factors, few studies have discussed the shape of the catchment area. This study focuses on analyzing the anisotropy of catchment areas and developing a sound methodology to generate them. Based on taxi global positioning system (GPS) data, this paper first proposes a data mining method to identify feeder taxi trips around subway stations. Then, a fan-shaped model is proposed and applied to Xi’an Metro Line 1 to generate catchment areas. The number and angle of fan areas are determined according to the spatial distribution characteristics of GPS points. Results show that the acceptable distance of the catchment area has significant differences in different directions. The average maximum acceptable distance of one station is 2.31 times the minimum. Furthermore, for feeder taxis, the overlap ratio of the catchment area is very high. Travelers in several places could choose several different stations during the travel. A multiple linear regression model was introduced to find the influencing factors, and the result shows the anisotropy of the catchment area is affected not only by neighboring subway stations, but also by the road network, distance from the city center, and so on.

A Sequential Pattern Mining Based Approach to Adaptively Detect Anomalous Paths in Floating Vehicle Trajectories

The detection of anomalous paths from floating vehicle trajectories plays an increasingly important role in dynamic path planning because of its ability to identify fraudulent transport behaviors, unexpected traffic accidents, and traffic restriction areas. Existing studies mostly emphasize the significant deviations based on global shape metrics and cannot manage trajectories with missing segments. To address these problems, this study proposes a new sequential pattern mining based method for detecting anomalous paths (SPDAP) in floating vehicle trajectories. First, discrete trajectory points were transformed into consecutive road segment sequences using map matching and essential empirical information. Focusing on these segment sequences, this method constructs a directed graph and calculates the conditional occurrence probabilities of multi-order sub-paths on the graph. On this basis, we designed a sequential pattern mining algorithm by estimating kernel density distributions to adaptively extract multi-order frequent sub-paths. Finally, these anomalous paths can be identified by matching the operations with frequent paths. Comparative experiments on simulated trajectories demonstrated that SPDAP could accurately, adequately, and adaptively detect anomalous paths. Additionally, using real-life taxi trajectories, we conducted semantic analysis by considering travel time and charges to further derive four patterns from the detected anomalous paths. These patterns are expected to support the identification of fraudulent taxi trips and the discovery of candidate driving routes with a preference for saving time or charge. In the future, we will focus on tracing the intrinsic causes of anomalous paths by introducing multisource information and performing real-time anomalous path detection using adaptive time intervals.

A Multi-Stage Optimisation Approach to Design Relocation Strategies in One-Way Car-Sharing Systems With Stackable Cars

One of the main operational challenges faced by the operators of one-way car-sharing systems is to ensure vehicle availability across the regions of the service areas with uneven patterns of rental requests. Fleet balancing strategies are required to maximise the demand served while minimising the relocation costs. However, the design of optimal relocation policies is a complex problem, and global optimisation solutions are often limited to very small network sizes for computational reasons. In this work, we propose a multi-stage decision support system for vehicle relocation that decomposes the general relocation problem into three independent decision stages to allow scalable solutions. Furthermore, we adopt a rolling horizon control strategy to cope with demand uncertainty. Our approach is highly modular and flexible, and we leverage it to design user-based, operator-based and robotic relocation schemes. Besides, we formulate the relocation problem considering both conventional cars and a new class of compact stackable vehicles that can be driven in a road train. We compare the proposed relocation schemes with two recognised benchmarks using a large data set of taxi trips in New York. Our results show that our approach is scalable and outperforms the benchmark schemes in terms of quality of service, vehicle utilisation and relocation efficiency. Furthermore, we find that stackable vehicles can achieve a relocation performance close to that of autonomous cars, even with a small workforce of relocators.

Uncovering Taxi Mobility Patterns Associated with the Public Transportation Shutdown Using Multisource Data in Washington, D.C.

The relationship between taxi travel patterns and public transportation disruption has not been extensively explored. In this study, we investigated the impact of public transportation disruption on the taxi mobility patterns during the metro shutdown in Washington, D.C.. Multiple data source, involving taxi trips, traffic analysis zone, and point of interest (POI) information, was collected to compare the taxi travel patterns before, during, and after the metro shutdown. The number, distance, and duration of taxi trips were found to be significantly higher during the metro shutdown; specifically, the number of taxi trips was found to be 19.8% larger. Furthermore, a POI auxiliary analysis was performed to investigate the variation in community structure during the disruption of public transport using the modularity maximization approach. The results of this study will be useful for the development of taxi scheduling strategies and traffic management.

Inferring the purposes of passenger trips from GPS records

A technological revolution has emerged in the context of mobility surveys with the widescale usability of wearable and onboard global positioning system (GPS) devices. Sri Lanka is also at the edge of utilizing these, in replacing the traditional methods. The earnestness' for this transition is supported by the deficiencies such as higher cost, higher nonresponses rates, over and under-reporting, and small sample sizes of traditional surveys. The activity that a passenger performs after a trip, or the purpose of a trip is a vital concern in transportation research as it is the reason behind the generation of travel demands. Hence, trip purpose inference from GPS data has become an important study area in this context [1]. Gong et al. [2] reviewing the existing trip purpose inference studies categorized the methodologies that had been used into three as rule-based, probabilistic, and machine learning-based. In this study, we utilized the GPS records of taxi trips from a popular service provider in the country as shown in Figure 1 in developing a suitable trip purpose inference approach.

Revealing Taxi Interaction Network of Urban Functional Area Units in Shenzhen, China

Characterizing the taxi travel network is of fundamental importance to our understanding of urban mobility, and could provide intellectual support for urban planning, traffic congestion, and even the spread of diseases. However, the research on the interaction network between urban functional area (UFA) units are limited and worthy of notice. Therefore, this study has applied the taxi big data to construct a travel flow network for the exploration of spatial interaction relationships between different UFA units in Shenzhen, China. Our results suggested that taxi travel behavior was more active in UFA units dominated by functions, including residential, commercial, scenic, and greenspace during weekends, while more active in UFA units dominated by industrial function during weekdays. In terms of daily average volume, the characteristics of spatial interaction between the various UFA types during weekdays and weekends were similar. During the morning peak period, the sink areas were mainly distributed in Futian District and Nanshan District, while during the evening peak period, the sink areas were mainly distributed in the southern part of Yantian District, the southwestern part of Longgang District, and the eastern part of Luohu District. The average daily taxi mobility network during weekdays showed a spatial pattern of “dense in the west and north, sparse in the south and east”, exhibiting significant spatial unevenness. Compared with weekdays, the daily taxi mobility network during weekends was more dispersed and the differences in node sizes decreased, indicating that taxi travel destinations were more diverse. The pattern of communities was more consistent with the administrative division during weekdays, indicating that taxi trips are predominantly within the districts. Compared with weekdays, the community pattern of network during weekends was clearly different and more in line with the characteristics of a small world network. The findings can provide a better understanding of urban mobility characteristics in Shenzhen, and provide a reference for urban transportation planning and management.

Development of predictive modeling and deep learning classification of taxi trip tolls

Several studies discussed the predictive modeling of deep learning in different applications such as classifying tissue features from microstructural data, Crude Oil Prices, mechanical constitutive behavior of materials, microbiome data, and mineral prospectively. Commercial navigation includes a wealth of trip-related data, including distance, expected journey time, and tolls that may be encountered along the way. Using a classification algorithm, it is possible to extract drop-off and pickup locations from taxi trip data and estimate if the tour would incur tolls. In this work, let’s use the classification learner to create classification models, compare their performance, and export the findings for additional study. The workflow for the classification learner is the same as for the regression learner. The purpose is to make predictions based on fresh data in order to see how well the model performs with new data. To train the model, it’s critical to separate the data set. The combined training and validation data is next pre-processed, which involves tasks such as cleaning and developing new features skills. Once the data has been prepared, it’s time to begin the supervised machine learning process and test a number of ways to identify the best model, such as the type of model that should be used, the important features, and the best parameters of the model to find the best fit for the considered data. The results of analyzing different predictive multiclass classification models with taxi trip tolls show that it is possible to use a machine learning-based model when we like to avoid road tolls depending on historical data on taxi trip tolls. The outcome of this study can help to expect road tolls from the drop-off and pickup locations of a taxi data

Clusters of Bicycle Taxi Operators and their Main Service Operation Patterns: Case Study of Quelimane, Mozambique

Bicycle taxi is a vital means of informal public transport service in most Sub-Saharan African cities, and for this reason, understanding who operates this service, and how they operate could help define initiatives to promote this service. This study considered clusters of bicycle taxi operators and their main service operation patterns. A survey was conducted among 105 regular bicycle taxi operators in Quelimane, Mozambique. Twostep cluster analysis identified homogeneous groups of bicycle taxi operators based on six socio-economic factors (age, income, education, household composition, bicycle ownership, and residence location). A Mann-Whitney U test was employed to compare pairs of clusters of bicycle taxi operators regarding a set of taxi services operation variables, such as the number of passengers carried daily, daily revenues, and service hours. Four clusters of bicycle taxi operators were identified which are, less-educated operators from large households (C1), educated migrants (C2), less-educated bicycle renters (C3), and young cyclists from small households (C4). When comparing differences in service operation patterns per cluster of bicycle taxi operators, the study showed that people in C1 produced fewer bicycle taxi trips than those in C2 and C4. For daily earnings, people in C2 earn more than those in C1 and C3. For service hours, individuals in C2 cycle long service hours when compared to those in C1, which could be harmful to their health. The result of this study could reorient bicycle taxi service promotional policies to make the service more sustainable.

Investigating functional consistency of mobility-related urban zones via motion-driven embedding vectors and local POI-type distributions

Urban morphology and human mobility are two sides of the complex mixture of elements that implicitly define urban functionality. By leveraging the emerging availability of crowdsourced data, we aim for novel insights on how they relate to each other, which remains a substantial scientific challenge. Specifically, our study focuses on extracting spatial-temporal information from taxi trips in an attempt on grouping urban space based on human mobility, and subsequently assess its potential relationship with urban functional characteristics in terms of local points-of-interest (POI) distribution. Proposing a vector representation of urban areas, constructed via unsupervised machine learning on trip data’s temporal and geographic factors, the underlying idea is to define areas as “related” if they often act as destinations of similar departing regions at similar points in time, regardless of any other explicit information. Hidden relations are mapped within the generated vector space, whereby areas are represented as points and stronger/weaker relatedness is conveyed through relative distances. The mobility-related outcome is then compared with the POI-type distribution across the urban environment, to assess the functional consistency of mobility-based clusters of urban areas. Results indicate a meaningful relationship between spatial-temporal motion patterns and urban distributions of a diverse selection of POI-type categorizations, paving the way to ideally identify homogenous urban functional zones only based on the movement of people. Our data-driven approach is intended to complement traditional urban development studies on providing a novel perspective to urban activity modeling, standing out as a reference for mining information out of mobility and POI data types in the context of urban management and planning.