Passenger Origin Destination Research Articles

Proposing a model for predicting passenger origin–destination in online taxi-hailing systems

Proposing a model for predicting passenger origin–destination in online taxi-hailing systems

Passenger origin–destination estimation in public transit using boarding count data

Passenger boarding and alighting data are critical for understanding travel patterns in transit systems, yet many systems only record boarding, making it difficult to track alighting. This study introduces a method to estimate alighting counts from boarding data using transit symmetry at a cluster level. Bus stops are grouped into clusters, assuming equal inflow and outflow, and a gravity model is calibrated to estimate an origin–destination (OD) matrix. Using data from three bus routes in Regina, Canada (2017), the method achieved a symmetry ratio of 60% across clusters and a mean structural similarity measure score of 0.84, validating the accuracy of the OD estimates. This approach provides a practical solution to the challenge of missing alighting data. While effective for single-mode transit systems, future research will expand its application to multi-modal systems and address factors like seasonal and land-use variations, offering a cost-effective alternative for OD estimation.

Deep Learning Model for Short-Term Origin–Destination Distribution Prediction in Urban Rail Transit Network Considering Destination Choice Behavior

Urban rail transit (URT) has emerged as a crucial mode of transportation in metropolitan areas. For the effective operation of expanding URT networks, accurate short-term origin–destination (OD) demand distribution predictions are essential. This study introduces a novel deep-learning-based model for predicting short-term OD distribution in extensive networks, taking destination choice behaviors into account. First, we perform a comprehensive analysis of station passenger flows and OD flows from both temporal and spatial dimensions. Then, we develop the origin–destination distribution prediction (ODDP) model, combining the destination choice model (DCM) with the deep learning model (DLM). The DCM aims to understand OD distribution patterns from a behavioral perspective by transforming real-time inflows into OD distributions. Meanwhile, the DLM, employing attention and convolution layers, effectively captures the intricate temporal and spatial dynamics of passenger flows. Our model is evaluated using data from the Guangzhou Metro network in China, showing significant enhancements in prediction accuracy, model interpretability, and overall robustness. The implementation of our model promises substantial benefits for the operational efficiency of URT systems.

Airline revenue management with segmented continuous pricing: methods and competitive effects

With the introduction of IATA’s New Distribution Capability (NDC), airlines will no longer be limited to discrete fare classes for their fare product distribution but could show fare quotes from continuous ranges to booking requests. NDC will also allow airlines to present different fare quotes to passengers from different demand segments as identified by the airline. In theory, airlines can better extract passenger willingness to pay, and thus, see gains in revenue, by offering segmented continuous fare quotes to different passengers requesting to book. This paper describes the revenue management (RM) methods for Segmented Continuous Pricing and examines their potential effects on airlines’ revenue through simulations in the Passenger Origin–Destination Simulator (PODS). We describe a class-based algorithm for continuous pricing, a straightforward extension from the traditional methods used with existing RM systems. Our simulation results show that in a calibrated scenario in which only one airline adopts Segmented Continuous Pricing and has an 80% accuracy in identifying business versus leisure passenger booking requests, the first-mover airline can see as much as a 17% revenue gain, at the expense of competitors. The revenue gains come primarily from the leisure passenger segment by offering lower fares than competitors closer to departure. The first-mover airline loses bookings but does not see losses in revenue from the business passenger segment. We also explore potential response strategies by the competing airlines. We discover that competitors can reverse the first-mover’s revenue gain by removing their fare restrictions while still using traditional RM methods. We conclude that although adopting Segmented Continuous Pricing is promising in theory, its gains in practice will depend heavily on the competitive situation and the responses made by competing airlines.

Bi-level model predictive control for metro networks: Integration of timetables, passenger flows, and train speed profiles

This paper deals with the train scheduling problem for metro networks taking into account time-dependent passenger origin–destination demands and train speed profiles. The aim is to adjust train schedules online according to time-dependent passenger demands so that passenger satisfaction and operational costs are jointly optimized. An extended passenger absorption model that explicitly includes time-dependent passenger origin–destination demands is developed, where the term “absorption” refers to passengers boarding trains. Then, the passenger absorption model is extended to a bi-level framework, where passenger demands and rolling stock availability are considered at the higher level, and detailed timetables and train speed profiles are included at the lower level. A bi-level model predictive control (MPC) approach is developed for the integrated problem. The optimization problems of both levels of the bi-level MPC approach can be converted into mixed-integer linear programming (MILP) problems, which enables us to solve them with existing MILP solvers. We then show that the recursive feasibility of both the higher-level and the lower-level optimization problems can be guaranteed. In this way, we can achieve real-time train scheduling for the metro system. Numerical experiments, based on real-life data from the Beijing metro network, illustrate the effectiveness of the extended passenger absorption model and the proposed bi-level MPC approach.

Improving service qualities of cyclic line plans considering heterogeneous passenger origin-destination (OD) flow groups

ABSTRACT Cyclic train services provide huge conveniences to passengers. This work aims to fix one vital promise in major line planning studies that passengers were homogeneous. In fact, passengers’ rail travel expectations are heterogeneous. But those heterogeneities show common needs correlating with travel distances and defined station levels. Therefore, we cluster passenger heterogeneities into divided OD groups, extend service qualities, and add more train line factors to our line planning model. Cyclic line plan from this joint optimisation not only receives service quality improvements but also provides multiple train service types to match clustered OD groups. Moreover, we allow passengers to be served by different train service types, deliver a high level of direct services, and maintain the minimum operation cost. We compare concluded line plans with the results of the generic line planning model and the real-world line plan to show our solution advantages.

Collaborative optimization for train stop planning and train timetabling on high-speed railways based on passenger demand.

In recent years, with increasing passenger travel demand, high-speed railways have developed rapidly. The stop planning and timetabling problems are the core contents of high-speed railway transport planning and have important practical significance for improving efficiency of passenger travel and railway operation Dong et al. (2020). This study proposes a collaborative optimization approach that can be divided into two phases. In the first phase, a mixed-integer nonlinear programming model is constructed to obtain a stop plan by minimizing the total passenger travel time. The constraints of passenger origin-destination (OD) demand, train capacity, and stop frequency are considered in the first phase. In the second phase, the train timetable is optimized after the stop plan is obtained. A multiobjective mixed-integer linear optimization model is formulated by minimizing the total train travel time and the deviation between the expected and actual departure times from the origin station for all trains. Multiple types of trains and more refined headways are considered in the timetabling model. Finally, the approach is applied to China's high-speed railway, and the GUROBI optimizer is used to solve the models in the above two stages. By analyzing the results, the total passenger travel time and train travel time decreased by 2.81% and 3.34% respectively. The proposed method generates a more efficient solution for the railway system.

Modeling and Efficient Passenger-Oriented Control for Urban Rail Transit Networks

Real-time timetable scheduling is an effective way to improve passenger satisfaction and to reduce operational costs in urban rail transit networks. In this paper, a novel passenger-oriented network model is developed for real-time timetable scheduling that can model time-dependent passenger origin-destination demands with consideration of a balanced trade-off between model accuracy and computation speed. Then, a model predictive control (MPC) approach is proposed for the timetable scheduling problem based on the developed model. The resulting MPC optimization problem is a nonlinear non-convex problem. In this context, the online computational complexity becomes the main issue for the real-time feasibility of MPC. To reduce the online computational complexity, the MPC optimization problem is therefore reformulated into a mixed-integer linear programming (MILP) problem. The resulting MILP problem is exactly equivalent to the original MPC optimization problem and can be solved very efficiently by existing MILP solvers, so that we can obtain the solution very fast and realize real-time timetable scheduling. Numerical experiments based on a part of Beijing subway network show the effectiveness and efficiency of the developed model and the MILP-based MPC method.

Scenario-Based MPC for Real-Time Passenger-Centric Timetable Scheduling of Urban Rail Transit Networks

Effective timetable scheduling strategies are essential for passenger satisfaction in urban rail transit networks. Most existing passenger-centric timetable scheduling approaches generate a timetable according to deterministic passenger origin-destination (OD) demands. As passenger OD demands in urban rail transit networks generally show a high level of uncertainty, an effective timetable scheduling approach should take the uncertain passenger flows into account to generate a reliable timetable. In this paper, a scenario-based model predictive control (SMPC) approach is presented to handle uncertain passenger flows based on a passenger absorption model, where uncertainties are captured by several representative scenarios according to historical data. In each SMPC step, the optimization problem for generating the timetable can be reformulated as a mixed-integer linear programming (MILP) problem, which can be efficiently solved using current MILP solvers. A probabilistic performance level can be then determined based on the performance of SMPC under the representative scenarios. Numerical experiments based on the Beijing subway network are conducted to evaluate the efficacy of the proposed approach.

Transforming Bus Service Planning Using Integrated Electronic Data Sources at NYC Transit

The installation of an Automatic Vehicle Location (AVL) system alongside existing Automated Fare Collection (AFC) data spurred development of an inferred bus boarding and alighting ridership model at New York City Transit (NYCT), allowing for 100% passenger origin-destination (O-D) data citywide. Analysis techniques that relied primarily on professional judgment due to lack of data were replaced by more sophisticated statistical techniques. This paper describes two case studies and the resulting service planning potential from having access to fully-integrated big data sources: a neighborhood-wide analysis of performance and ridership, where 100% data allowed planners to pinpoint specific, low-cost reroutes and stop changes to better serve riders, and identification of an optimal route split location for a long route with poor performance by minimizing passenger impact using modeled O-D data. In both examples, new data sources allowed for novel analysis throughout problem investigation as well as forecasting ridership and cost impacts of proposed service adjustments. As the agency’s ability to leverage these data improves, it will support Title VI obligations as well as performance monitoring.

Bus Passenger Origin-Destination Estimation and Related Analyses Using Automated Data Collection Systems

This research explores the application of archived data from Automated Data Collection Systems (ADCS) to transport planning with a focus on bus passenger travel behavior, including Origin-Destination (OD) inference, using London as a case study. It demonstrates the feasibility and ease of applying trip-chaining to infer bus passenger OD from smart card transactions and Automatic Vehicle Location (AVL) data and is the first known attempt to validate the results by comparing them with manual passenger survey data. With the inferred OD matrices, the variations of weekday and weekend bus route OD patterns are examined for planning purposes. Moreover, based on the inferred OD matrices and the AVL data, alighting times for bus passengers also can be estimated. Bus journey stages, therefore, can easily be linked. By comparing the interchange time and the connecting bus route’s headway, it provides a way to evaluate bus connections.

Short-Term Online Forecasting for Passenger Origin–Destination (OD) Flows of Urban Rail Transit: A Graph–Temporal Fused Deep Learning Method

Predicting short-term passenger flow accurately is of great significance for daily management and for a timely emergency response of rail transit networks. In this paper, we propose an attention-based Graph–Temporal Fused Neural Network (GTFNN) that can make online predictions of origin–destination (OD) flows in a large-scale urban transit network. In order to solve the key issue of the passenger hysteresis in online flow forecasting, the proposed GTFNN takes finished OD flow and a series of features, which are known or observable, as the input and performs multi-step prediction. The model is constructed from capturing both spatial and temporal characteristics. For learning spatial characteristics, a multi-layer graph neural network is proposed based on hidden relationships in the rail transit network. Then, we embedded the graph convolution into a Gated Recurrent Unit to learn spatial–temporal features. For learning temporal characteristics, a sequence-to-sequence structure embedded with the attention mechanism is proposed to enhance its ability to capture both local and global dependencies. Experiments based on real-world data collected from Chongqing’s rail transit system show that the metrics of GTFNN are better than other methods, e.g., the SMAPE (Symmetric Mean Absolute Percentage Error) score is about 14.16%, with a range from 5% to 20% higher compared to other methods.

Frequency-based path flow estimator for transit origin-destination trip matrices incorporating automatic passenger count and automatic fare collection data

A frequency-based path flow estimator is proposed to estimate the transit passenger origin–destination (OD) matrix in an urban congested transit network. The proposed model not only considers the effect of congestion under an equilibrium framework, but also benefits from being formulated as a single-level model with the route-section-based network representation. Multiple transit data sources are incorporated, including automatic passenger count, automatic fare collection and automatic vehicle location data. A path-based diagonalization approach embedded with an iterative balancing scheme is developed to solve the model. Case studies are conducted to demonstrate the features and applicability of the proposed model and algorithm.

Automation of Collection of Primary Data for Development of a Passenger Origin-Destination Trip Correspondence Matrix Based on Computer Vision and Neural Network Technologies

The origin-destination trip matrix is a fundamental characteristic of a transport network, and development of a reliable correspondence matrix is the most important task in organising passenger traffic. It is the basis on which the public transport route network of a city (region) is built and optimised.Currently, collection of initial information for construction of a travel correspondence matrix is carried out through field surveys comprising questionnaire surveys of the population; accounting for movement of passengers according to the coupons issued to them; checkers, tellers manually counting passengers in vehicle compartments; simple surveys of passengers. Besides, mathematical modelling is used based on statistical data on the number of residents in various districts of the city, employees in enterprises and students in educational institutions, as well as on available data on the characteristics of passenger traffic along certain routes. All these surveys are very expensive and are carried out once over few years; they give a large error, which is why decisions made on the basis of these data are far from being optimal.There are a lot of solutions in the software and hardware market that provide automated collection of data on passenger flows. They are based on the use of infrared sensors or of video recording. However, none of these systems provide information about the points of entry and exit of each passenger. The objective of this study was to develop methods for automating the collection of reliable information about passenger trips, that will be the base for building up-to-date and reliable passenger trip correspondence matrices. This task can be solved by constant monitoring of passengers’ trips with fixing places of entry and exit of each passenger.The study describes the possibility of creating software based on computer vision and artificial intelligence which will provide automation of collection of primary information about travel of each passenger from the place of boarding into the vehicle to exit from it, that is, automation of data generation to build a passenger trip correspondence matrix.

Research on origin‐destination travel demand prediction method of inter‐regional online taxi based on SpatialOD‐BiConvLSTM

In order to avoid empty or crowded trips of online taxis and shorten the distance between empty taxis and potential passengers, after adopting the Origin-Destination (OD) big data of taxis and considering geographic location information, cross-regional driving demand, as well as passenger origin demand and destination demand, a method combining SpatialOD and Bidirectional ConvLSTM (BiConvLSTM) neural network model is proposed to predict online taxi travel demand. The network model in the method takes into account the historical and future states of the data, and can extract the time and space characteristics of the data. A custom structural similarity loss function is used in the model training to measure the difference between the data. Through experiments on a large public taxi travel data set, the results show that the SMAPE error of this method is 23.83%, which is better than the existing OD demand forecasting methods and has strong practicability.

Non-Symmetric Spatial-Temporal Network for Bus Origin–Destination Demand Prediction

Urban public transport has become a preferred choice for alleviating traffic congestion. The bus passenger OD (origin–destination) demand prediction based on bus operational data is the key technology to realize urban intelligent transportation system. However, most of the existing bus OD demand prediction methods only considered regional passengers. The problem of the OD demand prediction based on historical OD matrices of bus lines is still not easy to implement, exceptionally, which is suitable for most of the urban bus lines. This paper presents a non-symmetric spatial-temporal network (NSTN) based on convolutional neural network (CNN) and convolutional long short-term memory (ConvLSTM) network to predict bus OD. NSTN contains the station spatial component (SSC) module and the spatial-temporal component (STC) module. SSC consists of two CNNs to learn the OD features and the DO (destination-origin) features, respectively. To make the prediction shift to the OD features, the non-symmetric input is designed. STC extracts spatial-temporal features based on ConvLSTM. Compared with other methods, NSTN has the best performance measured by symmetric mean absolute percentage error (SMAPE) and root mean square error (RMSE), where its SMAPE falls by 4.3 percentage points to 16.4 percentage points and RMSE falls by 23.1 percentage points to 69.9 percentage points. Experimental results on other bus lines show that NSTN has strong generalization ability.

The value of additional data for public transport origin–destination matrix estimation

Passenger origin–destination data is an important input for public transport planning. In recent years, new data sources have become increasingly common through the use of the automatic collection of entry counts, exit counts and link flows. However, collecting such data can be sometimes costly. The value of additional data collection hence has to be weighed against its costs. We study the value of additional data for estimating time-dependent origin–destination matrices, using a case study from the London Piccadilly underground line. Our focus is on how the precision of the estimated matrix increases when additional data on link flow, destination count and/or average travel distance is added, starting from origin counts only. We concentrate on the precision of the most policy-relevant estimation outputs, namely, link flows and station exit flows. Our results suggest that link flows are harder to estimate than exit flows, and only using entry and exit data is far from enough to estimate link flows with any precision. Information about the average trip distance adds greatly to the estimation precision. The marginal value of additional destination counts decreases only slowly, so a relatively large number of exit station measurement points seem warranted. Link flow data for a subset of links hardly add to the precision, especially if other data have already been added.

Approaches to Determining the Regular Transit Route Duplication Level

Issues of duplication of regular transit routes are of particular importance in the field of transport services provided to population and organisation of passenger transportation from the perspective of ensuring compliance with passengers needs for transportation and of the effects of route duplication on the technical, operational, and economic indicators of performance of these routes and the integral route network.In Russia duplication of regular routes within route networks emerged in the late 1990s – early 2000s in urban transit, other transit modes, and in interconnected transit. In the last decade, these routes have been increasingly subject to revision by local governments and executive bodies of federal constituent entities of the Russian Federation while solving transport planning problems and improving quality of transport services for the population.Evaluation of route duplication, as a rule, is carried out based on the route factor and the route duplication factor, the latter allows pairwise assessment of routes by the length of their overlapping segments.The objective of this article is to show incorrectness of the widespread technique and to present another approach that provides, in the author’s opinion, the correct interpretation of the method for determining the route duplication rate. Achieving this objective is based on methods of theoretical research in the field of organising passenger transportation.In development of this logic, the author proposed a new method for determining the route duplication factor using route adjacency factor, which considers directions and volumes of passenger origin-destination flows. Comparison of existing and proposed approaches is given using simple examples. The results of calculations have confirmed the different nature of factors and the absence of a direct relationship between the needs of passengers for transportation by public transport and the length of adjacent sections of routes. The conclusion is made about probable expediency of using the second (author’s) approach based on the route adjacency factor, which provides a correct solution to the stated transport planning problem. Besides, the possibility of using a new approach when performing diagnostics or designing route networks of different transport modes is shown both in relation to route matching and regarding their clusters and the entire route network.

Dynamic offer generation in airline revenue management

The New Distribution Capability and new retailing platforms will enable airlines to respond to shopping requests with bundled offers of flights and ancillary services, representing an evolution from traditional, flight-focused optimization. Assembling an attractive set of offers to display to customers therefore represents a new joint pricing and assortment optimization problem in airline revenue management. In this paper, we introduce an initial optimization approach for the selection and pricing of a la carte and bundled flight and ancillary offers. First, we propose a customer choice model that captures the impact of ancillary bundles on flight itinerary choice. We then calculate prices for each offer from a continuous range of price points and display the offer set that maximizes expected revenue for a given customer segment. We illustrate the approach using a single-flight, single-ancillary base case and discuss extensions to more complex environments. Tests in the Passenger Origin–Destination Simulator (PODS) show that dynamic offer generation (DOG) can increase net revenue when used by one or more airlines in a competitive network, assuming that the airlines are able to accurately segment incoming requests and estimate the average willingness-to-pay of each segment. We find that the majority of the revenue gains of DOG are due to competitive effects from the dynamic pricing of the flight component of the offer. The bundling mechanism of DOG is a secondary source of revenue gain that can be realized when customers take bundled ancillary services into account when choosing the flight. Our results provide insight for practitioners that are implementing offer optimization systems and processes. For example, in line with the previous literature on bundle pricing, we find that in transparent distribution channels an ancillary service should be bundled with the flight when the valuation for the ancillary is high or when its marginal cost of provision is low. We close by discussing the strategic and managerial implications of a move from traditional distribution strategies to a next-generation, offer-focused approach.

Continuous pricing algorithms for airline RM: revenue gains and competitive impacts

Traditionally, airlines have been limited to a set of fixed fare classes and, in turn, price points, to distribute their fare products. The advent of IATA’s new distribution capability (NDC) will soon enable airlines to quote any fare from a continuous range. In theory, such continuous pricing could increase revenues by extracting more of the consumer surplus, through its ability to offer more granular fares, closer to the customer’s willingness-to-pay (WTP). In this article, we describe several algorithms that lead to the quotation of a single fare from a continuous range. These algorithms either rely on traditional fare classes for the purpose of forecasting and optimization (class-based), or completely abandon the notion of fare classes, instead assuming different WTP distributions within each booking period prior to departure (classless). We describe how these algorithms build upon and differ from their traditional RM counterparts. Performance of these heuristics is then benchmarked against traditional class-based RM, and competitive impacts are analyzed when continuous pricing is adopted by one airline asymmetrically or both airlines symmetrically in a hypothetical 2-carrier network in the passenger origin–destination simulator (PODS). We find that continuous pricing is generally revenue-positive, and the revenue gains can be as high as 2.0% for the first-mover and reach up to 1.2% when both airlines adopt the new method. In addition, we show that these gains depend on the number of fare classes in the traditional fare structure used as a baseline, and that they are smaller under lower demand-to-capacity ratios.