Behavior of taxi customers in hailing vacant taxis: a nested logit model for policy analysis

This study models and examines the taxi customers' preferences for hailing vacant taxis on streets. A stated preference survey was conducted to randomly select and interview 1242 taxi customers at taxi stands and pedestrians on streets, who had experiences of taking taxis recently, about their choices under different given hypothetical scenarios. In total, 4968 observations were collected and used for developing the discrete choice models for the analysis. To account for the potential correlations among alternatives, two nested logit models are developed, calibrated, and compared with a standard multinomial logit model in the investigation. The results of likelihood ratio test demonstrate that one of the developed nested logit models is better than the standard multinomial logit model to describe the search behavior of taxi customers. The model results also show that the walking time to and the waiting time at the location for hailing taxis, the extra travel time to the destination because of local circulation for finding a way from the pickup location heading to a passenger's destination, as well as the taxi customers' perceptions for walking to and waiting at taxi stands were found as significant factors to influence their decisions. In addition, the results of market segmentation analysis illustrate the variations in taxi‐search strategies of taxi customers in different districts and regions. Some policy implications on introducing more taxi stands and improving the utilization rates of taxi stands are also discussed. We believe that the proposed models, findings, and discussion are useful for developing micro‐simulation models to evaluate the performance of road traffic networks with taxi services and developing simulation‐based optimization models to answer policy questions related to taxi services. Copyright © 2015 John Wiley & Sons, Ltd.     

Development of destination choice model with pairwise district-level constants using taxi GPS data

In this paper, a destination choice model with pairwise district-level constants is proposed for trip distribution based on a nearly complete OD trip matrix in a region. It is found that the coefficients are weakly identified in a destination choice model with pairwise zone-level constants. Thus, a destination choice model with pairwise district-level constants is then proposed and an iterative algorithm is developed for model estimation. Herein, the “district” means a spatial aggregation of a number of zones. The proposed model is demonstrated through simulation experiments. Then, destination choice models with and without pairwise district-level constants are estimated based on GPS data of taxi passenger trips collected during morning peak hours within the Inner Ring Road of Shanghai, China. The datasets comprise 504,187 trip records and a sample of 10,000 taxi trips for model development. The zones used in the study are actually 961 residents’ committees while the districts are 52 residential districts that are spatial aggregations and upper-level administrative units of residents’ committees. It is found that the estimated value of time dramatically drops after the involvement of district-level constants, indicating that the traditional model tends to overestimate the value of time when ignoring pairwise associations between two zones in trip distribution. The proposed destination choice model can ensure its predicted trip OD matrix to match the observed one at district level. Thus, the proposed model has potential to be widely applied for trip distribution under the situation where a complete OD trip matrix can be observed.     

Analysis of taxi demand and supply in New York City: implications of recent taxi regulations

This paper investigates temporal and weather-related variation in taxi trips in New York City. A taxi trip data-set with 147 million records covering 10 months of activity is used. It is shown that there are substantial variations in ridership, taxi supply, trip distance, and pickup frequency for different time periods and weather conditions. These variations, in turn, cause variations in driver revenues which is one of the main measures of taxi supply–demand equilibrium. The findings are then used to discuss the anticipated impacts of two recently enacted taxi regulation changes: the first fare increase since 2006 and the E-Hail pilot program which allows taxi hailing with smart phone applications. The fare increase is estimated to cause varying levels of revenue increase for different time periods. E-Hail apps are not expected to offer considerable improvements at all times, but rather when both adequate taxi supply and demand occur simultaneously.     

Optimizing the locations of electric taxi charging stations: A spatial–temporal demand coverage approach

Vehicle electrification is a promising approach towards attaining green transportation. However, the absence of charging stations limits the penetration of electric vehicles. Current approaches for optimizing the locations of charging stations suffer from challenges associated with spatial–temporal dynamic travel demands and the lengthy period required for the charging process. The present article uses the electric taxi (ET) as an example to develop a spatial–temporal demand coverage approach for optimizing the placement of ET charging stations in the space–time context. To this end, public taxi demands with spatial and temporal attributes are extracted from massive taxi GPS data. The cyclical interactions between taxi demands, ETs, and charging stations are modeled with a spatial–temporal path tool. A location model is developed to maximize the level of ET service on the road network and the level of charging service at the stations under spatial and temporal constraints such as the ET range, the charging time, and the capacity of charging stations. The reduced carbon emission generated by used ETs with located charging stations is also evaluated. An experiment conducted in Shenzhen, China demonstrates that the proposed approach not only exhibits good performance in determining ET charging station locations by considering temporal attributes, but also achieves a high quality trade-off between the levels of ET service and charging service. The proposed approach and obtained results help the decision-making of urban ET charging station siting.     

Solution methods for the taxi pooling problem

In Taiwan, taxi pooling is currently performed by some taxi companies using a trial-and-error experience-based method, which is neither effective nor efficient. There is, however, little in the literature on effective models and solution methods for solving the taxi pooling problem. Thus, in this study we employ network flow techniques and a mathematical programming method to develop a taxi pooling solution method. This method is composed of three models. First, a fleet routing/scheduling model is constructed to produce fleet/passenger routes and schedules. A solution algorithm, based on Lagrangian relaxation, a sub-gradient method and a heuristic to find the upper bound of the solution, is proposed to solve the fleet routing/scheduling model. Then, two single taxi-passenger matching models are constructed with the goals of decreasing number of passenger transfers and matching all passengers and taxis. These two taxi-passenger matching models are directly solved using a mathematical programming solver. For comparison with the solution method, we also develop another heuristic by modifying a heuristic recently proposed for solving a one-to-many taxi pooling problem. The performance of the solution method and the additional heuristic are evaluated by carrying out a case study using real data and suitable assumptions. The test results show that these two solution methods could be useful in practice.     

A data envelopment analysis model for evaluating the performance of taxi operators: a case study in Harbin,China

ABSTRACT

This paper measures the performance of taxi operators using data envelopment analysis (DEA) and the improvement of performance based on benchmarks. This includes basic DEA models and a cross efficiency model. In the present work, a sample set of 20 taxi operators are considered, and an evaluation indicator framework is constructed, which mainly involves the identification of decision-making units (DMUs) and the determination of input-output variables. The combined results of multiple DEA models are used to evaluate the performance of taxi operators in a case study of Harbin. The results highlight the significance of the combined application of multiple DEA models, especially a cross-efficiency model that can obtain more objective efficiency values. Finally, the benchmarking method is used to capture the achievements of best-performing operators and provide valuable improvement recommendations for inefficient operators. Conclusions provide a suitable basis for the operation and management of the taxi market.     

Accuracy of reinforcement learning algorithms for predicting aircraft taxi-out times: A case-study of Tampa Bay departures

Taxi-out delay is a significant portion of the block time of a flight. Uncertainty in taxi-out times reduces predictability of arrival times at the destination. This in turn results in inefficient use of airline resources such as aircraft, crew, and ground personnel. Taxi-out time prediction is also a first step in enabling schedule modifications that would help mitigate congestion and reduce emissions. The dynamically changing operation at the airport makes it difficult to accurately predict taxi-out time. In this paper we investigate the accuracy of taxi out time prediction using a nonparametric reinforcement learning (RL) based method, set in the probabilistic framework of stochastic dynamic programming. A case-study of Tampa International Airport (TPA) shows that on an average, with 93.7% probability, on any given day, our predicted mean taxi-out time for any given quarter, matches the actual mean taxi-out time for the same quarter with a standard error of 1.5 min. Also, for individual flights, the taxi-out time of 81% of them were predicted accurately within a standard error of 2 min. The predictions were done 15 min before gate departure. Gate OUT, wheels OFF, wheels ON, and gate IN (OOOI) data available in the Aviation System Performance Metric (ASPM) database maintained by the Federal Aviation Administration (FAA) was used to model and analyze the problem. The prediction accuracy is high even without the use of detailed track data.     

Modeling taxi drivers’ decisions for improving airport ground access: John F. Kennedy airport case

Taxi service is an important component of airport ground access, which affects the economic competitiveness of an airport and its potential positive impact on the surrounding region. Airports across the globe experience both taxi shortages and excesses due to various factors such as the airport’s proximity to the city center, timing and frequency of flights, and the fare structure. Since taxi drivers are independent entities whose decisions affect the taxi supply at airports, it is important to understand taxi drivers’ decision mechanisms in order to suggest policies and to maintain taxi demand and supply equilibrium at the airports. In this paper, New York City (NYC) taxi drivers’ decisions about airport pick-ups or cruising for customers at the end of each trip is modeled using logistic regression based on a large taxi GPS dataset. The presented approach helps to quantify the potential impacts of parameters and to rank their influence for policy recommendations. The results reveal that spatial variables (mainly related to proximity) have the highest impact on taxi drivers’ airport pickup decisions, followed by temporal, environmental and driver-shift related variables. Along with supplementary information from unstructured taxi driver interviews, the model results are used to suggest policies for the improvement of John F. Kennedy (JFK) airport’s ground access and passenger satisfaction, i.e. the implementation of taxi driver frequent airport server punch cards and a time-specific ride share program.     

Bus travel time prediction using a time-space discretization approach

The accuracy of travel time information given to passengers plays a key role in the success of any Advanced Public Transportation Systems (APTS) application. In order to improve the accuracy of such applications, one should carefully develop a prediction method. A majority of the available prediction methods considered the variation in travel time either spatially or temporally. The present study developed a prediction method that considers both temporal and spatial variations in travel time. The conservation of vehicles equation in terms of flow and density was first re-written in terms of speed in the form of a partial differential equation using traffic stream models. Then, the developed speed based equation was discretized using the Godunov scheme and used in the prediction scheme that was based on the Kalman filter. From the results, it was found that the proposed method was able to perform better than historical average, regression, and ANN methods and the methods that considered either temporal or spatial variations alone. Finally, a formulation was developed to check the effect of side roads on prediction accuracy and it was found that the additional requirement in terms of location based data did not result in an appreciable change in the prediction accuracy. This clearly demonstrated that the proposed approach based on using vehicle tracking data is good enough for the considered application of bus travel time prediction.     

An agent‐based simulation model to assess the impacts of introducing a shared‐taxi system: an application to Lisbon (Portugal)

This paper presents a new concept of urban shared‐taxi services. The proposed system has a new organisational design and pricing scheme that aims to use the capacity in traditional taxi services in a more efficient way. In this system, a taxi acting in ‘sharing’ mode offers lower prices to its clients, in exchange for them to accept sharing the vehicle with other persons who have compatible trips (time and space). The paper proposes and tests an agent‐based simulation model in which a set of rules for space and time matching between a request of a client and the candidate shared taxis is identified. It considers that the client is only willing to accept a maximum deviation from his or her direct route and establishes an objective function for selecting the best candidate taxi. The function considers the minimum travel time combination of pickup and drop‐off of all the pool of clients sharing each taxi while allowing to establish a policy of bonuses to competing taxis with certain number of occupants. An experiment for the city of Lisbon is presented with the objectives of testing the proposed simulation conceptual model and showing the potential of sharing taxis for improving mobility management in urban areas. Results show that the proposed system may lead to significant fare and travel time savings to passengers, while not jeopardising that much the taxi revenues. Copyright © 2014 John Wiley & Sons, Ltd.     

Procedures for planning multi-leg journeys with fixed-route and demand-responsive passenger transport services

This paper describes journey-planning procedures designed for use in a traveller information system covering fixed-schedule and demand-responsive public transport modes. The task is to construct a sequence of journey-legs to meet a traveller’s requirements with the least possible generalised cost, subject to time-window and other constraints. A journey may be carried out in a single leg by walking or by taking a taxi all the way from the origin to the destination, or by a sequence of one or more legs carried by public transport services sandwiched between walked segments connecting an initial pickup and final setdown stop. The public transport services may include fixed-route modes such as bus and train, and demand-responsive services running between fixed points. The main planning procedures are a high-level request-broker and a branch and bound procedure to handle multi-legged journeys; the request-broker also invokes a fleet-scheduling module to obtain bookings on demand-responsive services. The paper describes planning conditions, the planning procedures, and reduction techniques that are used to obtain acceptable computational performance. Tests with simulated demand suggest that the procedures are well suited for use in a real-time traveller information system.     

Survey and empirical evaluation of nonhomogeneous arrival process models with taxi data

Arrival processes are important inputs to many transportation system functions, such as vehicle prepositioning, taxi dispatch, bus holding strategies, and dynamic pricing. We conduct a comprehensive survey of the literature which shows that many transport systems employ basic homogeneous arrival process models or static nonhomogeneous processes. We conduct an empirical experiment to compare five state of the art arrival process short term prediction models using a common transportation system data set: New York taxi passenger pickups in 2013. Pickup data is split between 672 observations for model estimation and 96 observations for validation. From our experiment, we obtain evidence to support a recent model called FM‐IntGARCH, which is able to combine the benefits of both time series models and discrete count processes. Using a set of seven performance metrics from the literature, FM‐IntGARCH is shown to outperform the offline models—seasonal factor method, piecewise linear model—as well as the online models—ARIMA, Gaussian Cox process. Implications for operating data‐driven “smart” transit systems and urban informatics are discussed. Copyright © 2016 John Wiley & Sons, Ltd.     

On the use of Lagrangian observations from public transport and probe vehicles to estimate car space-mean speeds in bi-modal urban networks

The Macroscopic Fundamental Diagram (MFD) has been recognized as a powerful framework to develop network-wide control strategies. Recently, the concept has been extended to the three-dimensional MFD, used to investigate traffic dynamics of multi-modal urban cities, where different transport modes compete for, and share the limited road infrastructure. In most cases, the macroscopic traffic variables are estimated using either loop detector data (LDD) or floating car data (FCD). Taking into account that none of these data sources might be available, in this study we propose novel estimation methods for the space-mean speed of cars based on: (i) the automatic vehicle location (AVL) data of public transport where no FCD is available; and (ii) the fused FCD and AVL data sources where both are available, but FCD is not complete. Both methods account for the network configuration layout and the configuration of the public transport system. The first method allows one to derive either uni-modal or bi-modal macroscopic fundamental relationships, even in the extreme cases where no LDD nor FCD exist. The second method does not require a priori knowledge about FCD penetration rates and can significantly improve the estimation accuracy of the macroscopic fundamental relationships. Using empirical data from the city of Zurich, we demonstrate the applicability and validate the accuracy of the proposed methods in real-life traffic scenarios, providing a cross-comparison with the existing estimation methods. Such empirical comparison is, to the best of our knowledge, the first of its kind. The findings show that the proposed AVL-based estimation method can provide a good approximation of the average speed of cars at the network level. On the other hand, by fusing the FCD and AVL data, especially in case of sparse FCD, it is possible to obtain a more representative outcome regarding the performance of multi-modal traffic.     

Prediction of arrival times of freight traffic on US railroads using support vector regression

Variability of travel times on the United States freight rail network is high due to large network demands relative to infrastructure capacity, especially when traffic is heterogeneous. Variable runtimes pose significant operational challenges if the nature of runtime variability is not predictable. To address this issue, this article proposes a data-driven approach to predict estimated times of arrival (ETAs) of individual freight trains, based on the properties of the train, the properties of the network, and the properties of potentially conflicting traffic on the network. The ETA prediction problem from an origin to a destination is posed as a machine learning regression problem and solved using support vector regression trained and cross validated on over two years of detailed historical data for a 140 mile section of track located primarily in Tennessee, USA. The article presents the data used in this problem and details on feature engineering and construction for predictions made across the full route. It also highlights findings on the dominant sources of runtime variability and the most predictive factors for ETA. Improvement results for ETA exceed 21% over a baseline prediction method at some locations and average 14% across the study area.     

A gradient boosting logit model to investigate driver’s stop-or-run behavior at signalized intersections using high-resolution traffic data

Driver’s stop-or-run behavior at signalized intersection has become a major concern for the intersection safety. While many studies were undertaken to model and predict drivers’ stop-or-run (SoR) behaviors including Yellow-Light-Running (YLR) and Red-Light-Running (RLR) using traditional statistical regression models, a critical problem for these models is that the relative influences of predictor variables on driver’s SoR behavior could not be evaluated. To address this challenge, this research proposes a new approach which applies a recently developed data mining approach called gradient boosting logit model to handle different types of predictor variables, fit complex nonlinear relationships among variables, and automatically disentangle interaction effects between influential factors using high-resolution traffic and signal event data collected from loop detectors. Particularly, this research will first identify a series of related influential factors including signal timing information, surrounding traffic information, and surrounding drivers’ behaviors using thousands drivers’ decision events including YLR, RLR, and first-to-stop (FSTP) extracted from high-resolution loop detector data from three intersections. Then the research applies the proposed data mining approach to search for the optimal prediction model for each intersection. Furthermore, a comparison was conducted to compare the proposed new method with the traditional statistical regression model. The results show that the gradient boosting logit model has superior performance in terms of prediction accuracy. In contrast to other machine learning methods which usually apply ‘black-box’ procedures, the gradient boosting logit model can identify and rank the relative importance of influential factors on driver’s stop-or-run behavior prediction. This study brings great potential for future practical applications since loops have been widely implemented in many intersections and can collect data in real time. This research is expected to contribute to the improvement of intersection safety significantly.     

Cellpath: Fusion of cellular and traffic sensor data for route flow estimation via convex optimization

A new convex optimization framework is developed for the route flow estimation problem from the fusion of vehicle count and cellular network data. The issue of highly underdetermined link flow based methods in transportation networks is investigated, then solved using the proposed concept of cellpaths for cellular network data. With this data-driven approach, our proposed approach is versatile: it is compatible with other data sources, and it is model agnostic and thus compatible with user equilibrium, system-optimum, Stackelberg concepts, and other models. Using a dimensionality reduction scheme, we design a projected gradient algorithm suitable for the proposed route flow estimation problem. The algorithm solves a block isotonic regression problem in the projection step in linear time. The accuracy, computational efficiency, and versatility of the proposed approach are validated on the I-210 corridor near Los Angeles, where we achieve 90% route flow accuracy with 1033 traffic sensors and 1000 cellular towers covering a large network of highways and arterials with more than 20,000 links. In contrast to long-term land use planning applications, we demonstrate the first system to our knowledge that can produce route-level flow estimates suitable for short time horizon prediction and control applications in traffic management. Our system is open source and available for validation and extension.     

Optimizing charging station locations for urban taxi providers

Recently, electric vehicles are gaining importance which helps to reduce dependency on oil, increases energy efficiency of transportation, reduces carbon emissions and noise, and avoids tail pipe emissions. Because of short daily driving distances, high mileage, and intermediate waiting time, fossil-fuelled taxi vehicles are ideal candidates for being replaced by battery electric vehicles (BEVs). Moreover, taxi BEVs would increase visibility of electric mobility and therefore encourage others to purchase an electric vehicle. Prior to replacing conventional taxis with BEVs, a suitable charging infrastructure has to be established. This infrastructure consists of a sufficiently dense network of charging stations taking into account the lower driving ranges of BEVs.In this case study we propose a decision support system for placing charging stations in order to satisfy the charging demand of electric taxi vehicles. Operational taxi data from about 800 vehicles is used to identify and estimate the charging demand for electric taxis based on frequent origins and destinations of trips. Next, a variant of the maximal covering location problem is formulated and solved to satisfy as much charging demand as possible with a limited number of charging stations. Already existing fast charging locations are considered in the optimization problem. In this work, we focus on finding regions in which charging stations should be placed rather than exact locations. The exact location within an area is identified in a post-optimization phase (e.g., by authorities), where environmental conditions are considered, e.g., the capacity of the power network, availability of space, and legal issues.Our approach is implemented in the city of Vienna, Austria, in the course of an applied research project that has been conducted in 2014. Local authorities, power network operators, representatives of taxi driver guilds as well as a radio taxi provider participated in the project and identified exact locations for charging stations based on our decision support system.     

Dynamic modeling and control of taxi services in large-scale urban networks: A macroscopic approach

Taxis are increasingly becoming a prominent mobility mode in many major cities due to their accessibility and convenience. The growing number of taxi trips and the increasing contribution of taxis to traffic congestion are cause for concern when vacant taxis are not distributed optimally within the city and are unable to find unserved passengers effectively. A way of improving taxi operations is to deploy a taxi dispatch system that matches the vacant taxis and waiting passengers while considering the search friction dynamics. This paper presents a network-scale taxi dispatch model that takes into account the interrelated impact of normal traffic flows and taxi dynamics while optimizing for an effective dispatching system. The proposed model builds on the concept of the macroscopic fundamental diagram (MFD) to represent the dynamic evolution of traffic conditions. The model considers multiple taxi service firms operating in a heterogeneously congested city, where the city is assumed to be partitioned into multiple regions each represented with a well-defined MFD. A model predictive control approach is devised to control the taxi dispatch system. The results show that lack of the taxi dispatching system leads to severe accumulation of unserved taxi passengers and vacant taxis in different regions whereas the dispatch system improves the taxi service performance and reduces traffic congestion by regulating the network towards the undersaturated condition. The proposed framework demonstrates sound potential management schemes for emerging mobility solutions such as fleet of automated vehicles and demand-responsive transit services.     

Comparative tests of multimodal traffic equilibrium methods

In this study we proceeded to test the relative efficiencies of the two main computational techniques now available for calculating the traffic equilibrium in multimodal networks, namely, the relaxation and projection methods. Since both of these methods solve at each step a mathematical programming problem, we first set out to test the Frank-Wolfe algorithm and the Dafermos-Sparrow algorithm and found the latter to be the more efficient algorithm. As expected, this was also the case when these algorithms were used to solve the mathematical programming problem at each step of the relaxation method. We then investigated how different versions of the projection method enhance its performance. Subsequently, we proceeded to our main goal to compare the best projection method with the relaxation method. We tested multimodal networks with three different classes of monotone travel cost functions and found that the form of the travel cost functions affects the performance of the two basic techniques available for computing the multimodal equilibrium.