Origin-destination Estimation Research Articles

A combined method to forecast and estimate traffic demand in urban networks

This paper presents a combined method for short-term forecasting of detector counts in urban networks and subsequent traffic demand estimation using the forecasted counts as constraints to estimate origin–destination (OD) flows, route and link volumes. The method is intended to be used in the framework of an adaptive traffic control strategy with consecutive optimization intervals of 15min. The method continuously estimates the forthcoming traffic demand that can be used as input data for the optimization. The forecasting uses current and reference space–time-patterns of detector counts. The reference patterns are derived from data collected in the past. The current pattern comprises all detector counts of the last four time intervals. A simple but effective pattern matching is used for forecasting. The subsequent demand estimation is based on the information minimization model that has been integrated into an iterative procedure with repeated traffic assignment and matrix estimation until a stable solution is found. Some enhancements including the improvement of constraints, redundancy elimination of these constraints and a travel time estimation based on a macroscopic simulation using the Cell Transmission Model have been implemented. The overall method, its modules and its performance, which has been assessed using artificially created data for a real sub-network in Hannover, Germany, by means of a microsimulation with Aimsun NG, are presented in this paper.

A hierarchical optimization problem: Estimating traffic flow using Gamma random variables in a Bayesian context

In this paper a hierarchical optimization problem generated by a Bayesian method to estimate origin–destination matrices, based on Gamma models, is given. The problem can be considered as a system of equations in which three of them are optimization problems: (1) a Wardrop minimum variance (WMV) assignment model, which is used to derive the route choice probabilities, (2) a least squares problem, used to obtain the OD sample data, and (3) a maximum likelihood problem to estimate the posterior modes. A multi-level iterative approach is proposed to solve the multi-objective problem that converges in a few iterations. Finally, two examples of applications are used to illustrate the proposed methods and procedures, a simple and the medium size Ciudad Real networks. A comparison with existing techniques, which provide similar flows, seems to validate the proposed methods.

Accuracy of On-site Consistent Origin-Destination Estimation Method for Measuring the Actual Numbers of Consumer Shop-around Movements within the City Center of Hanoi, Vietnam

Accuracy of On-site Consistent Origin-Destination Estimation Method for Measuring the Actual Numbers of Consumer Shop-around Movements within the City Center of Hanoi, Vietnam

고속도로 IC간 교통수요 추정과 이를 통한 교통정보 제공 알고리즘 개발

The objective of VMS(Variable Message Sign) is to provide the traffic information downstream to drivers upstream so that they can choose their routes or expect the travel time to arrive the destination. Because there is not enough time and space to show the message, VMS message should be selected carefully. However, the message of VMS has been simply selected among the pre-designed message sets based on the priority rule of events. If the traffic demand between origin and destination is identified along the freeway, message can be selected to provide the information of a route that more drivers will use. In this study, a time sliced OD(Origin/Destination) estimation method will be developed using the detector information of the on-ramp, exit ramp, and the main lanes. And the strategy of a priority rule of message was planned.

Dynamic OD estimation using three phase traffic flow theory

AbstractAdvanced Transportation Management and Information Systems (ATMIS) can use dynamic origin–destination (OD) demand models to make short‐term predictions regarding developments in traffic states. However, existing dynamic OD prediction models do not achieve this reliably for two main reasons. First, this is a bi‐level system that consists of a traffic flow process at the lower level and a dynamic OD process at the upper level. Due to the inherent non‐convexity of bi‐level systems, it is difficult to guarantee that any calculated solution is globally optimal. In this paper, we propose a new traffic flow model that uses real‐time traffic data, such as traffic flows, speed and occupancy, collected from vehicle detectors, to address the difficulties that arise in existing bi‐level programming formulations. Second, in order to estimate a dynamic OD demand between on and off‐ramps on the freeways, a traffic flow model is needed to estimate the proportion of traffic moving between them. In this paper, we present a dynamic traffic estimation model based on Kerner's 1 three‐phase traffic theory, which represents the complexity of traffic phenomena based on phase transitions between free‐flow, synchronized flow and moving jam phases, and on their complex nonlinear spatio‐temporal features. The present model explains and estimates traffic congestion in terms of speed breakdown, phase transition and queue propagation. We show how a genetic algorithm can be used to solve this to estimate dynamic OD flows and the associated link, on and off‐ramp flows during each time interval using traffic data collected from vehicle detection systems implemented on Korean freeways. Copyright © 2010 John Wiley & Sons, Ltd.

Transit Stop-Level Origin–Destination Estimation through Use of Transit Schedule and Automated Data Collection System

As fare and data collection technology has developed, the resolution of collected data has reached the level of the individual traveler in investigations of transit passenger behavior. This paper investigates the use of these data to estimate passenger origins and destinations at the level of individual stops. Because of a lack of information from the fare collection system, researchers still need some estimate of passengers' alighting stops to complete each passenger trip chain on a specific day. Automated fare collection (AFC) and automated vehicle location (AVL) systems are the inputs to the estimation. Instead of typical AVL data, the paper proposes two models to estimate the alighting stop; both consider passenger trip chaining by using AFC data, transit schedule data (Google's General Transit Feed Specification), and automated passenger counter (APC) data. The paper validates the model by comparing the output to APC data with vehicle location data (APC-VL) and performs sensitivity analyses on several parameters in the models. To detect transfer trips, the new models propose a submodel that takes into account the effect of service headway in addition to some typical transfer time thresholds. Another contribution of this study is the relative relaxation of the search in finding the boarding stops, which enables the alternative algorithm to detect and fix possible errors in identification of the boarding stop for a transaction. As a result, the paper provides algorithms for the proposed models and sensitivity analysis for several predefined scenarios. The results are based on data and observed bus passenger behavior in the Minneapolis–Saint Paul, Minnesota, area.

The estimation of a time-dependent OD trip table with vehicle trajectory samples

In this paper we discuss a dynamic origin–destination (OD) estimation problem that has been used for identifying time-dependent travel demand on a road network. Even though a dynamic OD table is an indispensable data input for executing a dynamic traffic assignment, it is difficult to construct using the conventional OD construction method such as the four-step model. For this reason, a direct estimation method based on field traffic data such as link traffic counts has been used. However, the method does not account for a logical relationship between a travel demand pattern and socioeconomic attributes. In addition, the OD estimation method cannot guarantee the reliability of estimated results since the OD estimation problem has a property named the ‘underdetermined problem.’ In order to overcome such a problem, the method developed in this paper makes use of vehicle trajectory samples with link traffic counts. The new method is applied to numerical examples and shows promising capability for identifying a temporal and spatial travel demand pattern.

A Neural Network Approach to Motorway OD Matrix Estimation from Loop Counts

A method has been developed to estimate Origin Destination (OD) matrices using a neural network (NN) model and loop traffic data collected from a UK motorway site (M42) as the primary input. The estimated ODs were validated against matched vehicle number plate data derived from the ANPR (Automatic Number Plate Recognition) cameras which were installed at all the slip roads between junctions 3a and 7 of the motorway. Key research questions were: whether it is realistic to use the full loop data, whether particular features of the data influenced modelling success, whether data transformation could improve modelling performance through variance stabilization and whether individual ODs should be estimated separately or simultaneously. The method has been shown to work well and the best results were obtained using a square root transformation of the training data and individual models for each OD.

Empirical Study of Assignment-based Dynamic Origin-Destination Estimation and Predictions

Empirical Study of Assignment-based Dynamic Origin-Destination Estimation and Predictions

Time Dependent Origin-destination Estimation from Traffic Count without Prior Information

Existing research works on time-dependent origin-destination (O-D) estimation focus on the surveillance data usually assume the prior information of the O-D matrix (or transition matrix) is known (or at least partially known). In this paper, we relax such assumption by combining Gibbs sampler and Kalman filter in a state space model. A solution algorithm with parallel chain convergence control is proposed and implemented. To enhance its efficiency, a parallel structure is suggested with efficiency and speedup demonstrated using PC-cluster. Two numerical examples (one for Taipei Mass Rapid Transit network and the other for Taiwan Area National Freeway network) are included to show the proposed model could be effective of time-dependent origin-destination estimation.

A Markov Chain Monte Carlo-Based Origin Destination Matrix Estimator that is Robust to Imperfect Intelligent Transportation Systems Data

A method for estimating the Origin Destination (OD) split proportion matrix based on the observed traffic volume count data such as those from Intelligent Transportation System (ITS) is presented in this article. The nature of the ITS data, which frequently contains erroneous observations or missing values, requires that the procedure (1) is resistant to outlying errors, (2) can cope with missing values, and (3) can yield uncertainty estimates so that the reliability of the OD estimator can be assessed. The main goal of this article is to develop a robust estimation procedure that can handle both outliers and missing values and provide the interval estimates for the OD split proportion matrix. To accommodate outlying observations, a heavy-tailed error distribution (i.e., the t-distribution with low degrees of freedom) is assumed as opposed to assuming a Gaussian error distribution. Because the problem is intractable using traditional analytical techniques, a modern statistical computational technique, known as the Markov Chain Monte Carlo (MCMC) method, is employed. By using the MCMC method, the estimation of the split proportion matrix, interval estimation, and imputation of missing values can be done simultaneously. The main advantages of the new method are that (1) an extra step to identify outliers in data cleaning can be avoided, and (2) that partial observations (e.g., an observation containing missing values only at a few destinations) can be saved and utilized for OD estimation. In effect, the new OD estimation process effectively uses all ITS data in the OD estimation process. In addition, the method can provide statistically valid uncertainty estimates (interval estimates) for the estimated OD matrix even when there are outliers and/or missing values. The use of interval estimates in ITS applications has been limited in practice. However, using reasonable upper and lower bound estimates, rather than the mean estimate, will give more flexibility to ITS operators. The new MCMC method for OD estimation is evaluated using simulated data and observed ITS data from a test bed in San Antonio, Texas, that has been instrumented with inductance loops.

Vehicle Detector Deployment Strategies for the Estimation of Network Origin–Destination Demands Using Partial Link Traffic Counts

Making valid inferences about network origin-destination (OD) demands from limited link traffic count data requires a carefully structured data-collection strategy. Determination of the required number of vehicle detectors (VDs) and their installation in strategic locations for OD estimation purposes is essentially a network location problem. In this paper, a mathematical programming framework based on a user equilibrium assumption was developed to determine the most desirable locations for the deployment of VDs. A linearly independent model was also developed to deal with the network location problem. Proposed models were evaluated under different network sizes and varying OD demand levels. Numerical results are explained. These findings can beneficially contribute to the preparation of a desirable VD deployment plan in a general network.

A column generation algorithm for the estimation of origin–destination matrices in congested traffic networks

The general problem of estimating origin–destination (O–D) matrices in congested traffic networks is formulated as a mathematical programme with equilibrium constraints, referred to as the demand adjustment problem (DAP). This approach integrates the O–D matrix estimation and the network equilibrium assignment into one process. In this paper, a column generation algorithm for the DAP is presented. This algorithm iteratively solves a deterministic user equilibrium model for a given O–D matrix and a DAP restricted to the previously generated paths, whose solution generates a new O–D trip matrix estimation. The restricted DAP is formulated via a single level optimization problem. The convergence on local minimum of the proposed algorithm requires only the continuity of the link travel cost functions and the gauges used in the definition of the DAP. In order to analyze the convergence and performance of the proposed algorithm, various numerical tests were carried out on small scale problems.

Sensor Locations for Reliable Travel Time Prediction and Dynamic Management of Traffic Networks

State estimation and prediction models rely on the quality and the dissemination of the available traffic data. In complex networks, such as large urban areas, the position of traffic counts (i.e. loop detectors) is critical for such models. Nevertheless, sensor locations are typically chosen for other objectives (e.g., monitoring bottleneck sections, estimating origin–destination flows). As a result, link state estimations and predictions are often found to be inaccurate. A new approach to the sensor location problem was proposed; the objective was optimizing the position of traffic counts for reliable state estimation and prediction at complex networks. To do so, the maximum possible relative error, used in past approaches, was reformulated to minimize the error in link traffic states. By doing so, more reliable predictions of partial as well as whole network travel times can be obtained. Computing the new maximum possible relative error (MPRE) requires two extra inputs—the specification of a state estimation model and the specification of the network traffic variability—to set the boundaries of the solution space. For large complex networks, the latter information might be very difficult to obtain. As an alternative, a simple solution algorithm to the problem was proposed that uses link flow and travel time correlations between links to select, in sequence, the most representative locations in the network. The algorithm was tested on a toy network, which showed that it succeeds in catching a substantially larger percentage of link flows with respect to a classical approach.

Two-dimensionally constrained disaggregate trip generation, distribution and mode choice model: Theory and application for a Swiss national model

The Swiss federal government has asked the IVT, ETH Zürich in collaboration with the TU Dresden and Emch + Berger, Zürich to estimate origin–destination matrices by mode and purpose for the year 2000. The complex zoning system employing about 3000 zones required an algorithm which is fast, but also able to face generation, distribution and mode choice simultaneously. The EVA algorithm developed by Lohse et al. [Lohse, D., Teichert, H., Dugge, B., Bachner, G., 1997. Ermittlung von Verkehrsströmen mit n-linearen Gleichungssystemen unter Beachtung von Nebenbedingungen einschließlich Parameterschätzung (Verkehrsnachfragemodellierung: Erzeugung, Verteilung, Aufteilung). Schriftenreihe des Instituts für Verkehrsplanung und Straßenverkehr, H. 5/1997, Fakultät Verkehrswissenschaften “Friedrich List”, Technische Universität Dresden] was adapted for this purpose. The key properties of the algorithm are a disaggregate description of the demand, and its use of appropriate logit-type models for the demand distribution, while maintaining the known marginal distributions of the matrices generated. The algorithm calculates trip production and attractions by zone using activity pairs. The combined destination and mode choice models are estimated for the different traveller types and activity pairs. The paper derives and describes for the first time the EVA algorithm in English, including the solution method used. Second, it summarises the results of choice model estimation providing generalised cost elasticities of demand by purpose and traveller type. Third, it discusses the quality of the results by assessing the structure of the matrix against actual census data for road and rail traffic.

Time-dependent origin-destination estimation: Genetic algorithm-based optimization with updated assignment matrix

An estimation of origin-destination (OD) demand matrix is one of key elements to ensure the success in traffic modeling analysis. Even with widely deployed traffic sensors and advanced computational technologies, an estimation of time-dependent OD matrices is still a key barrier for the implementation of dynamic traffic assignment as well as simulation-based traffic modeling analysis. This paper proposes an improvement of existing time-dependent OD estimation method by updating assignment matrix at each step of OD estimation process and quantifies benefits and costs of doing so. The results from a case study with Florian network showed that the estimated OD flows from the proposed GA-based method with updated assignment matrix reduced the sum of square errors by 40% when compared with the OD flows from the DynaMIT OD estimation method with fixed assignment matrix, one of the most commonly used OD estimation methods. However, the proposed method would require significantly higher computational time than the traditional DyanMIT OD estimation method.

Genetically-optimized origin-destination estimation (GOODE) model: application to regional commodity movements in Ontario

The problem of origin-destination (O-D) matrix estimation has attracted significant research attention in the past few decades. This paper proposes a novel approach to estimate a regional freight O-D matrix using different data sources. The genetically optimized origin-destination estimation (GOODE) model takes advantage of the genetic algorithm's (GA) global search procedure to find the O-D matrix that is associated with the minimum deviation between estimated and observed data values. The GOODE-commodity model, an extension of the GOODE model, estimates the freight O-D matrix by interfacing GOODE with a trip generation model based on input-output data. The GOODE model and its extension bring together national input-output data, truck survey data, a global searching method, and a GIS platform for data manipulation. This paper presents the GOODE model structure, a prototypical numerical example, a benchmarking exercise with an existing O-D estimation model, and a real-world application of the GOODE-commodity model for a case study of commodity movements in Ontario. Avenues for future research are also addressed.Key words: origin-destination (O-D) matrix estimation, truck transportation modelling, input-output, Ontario, genetic algorithm.

System Wide Data Quality Control of Inductance Loop Data Using Nonlinear Optimization

One of the main criticisms and concerns over the years about inductive loop detector (ILD) data is systematic errors in the data associated with noncatastrophic malfunctioning of the devices. Most of the current approaches check for ILD data accuracy at individual locations. However, for an end application that requires data from neighboring locations, such as for travel time or origin-destination estimation, data quality control at a system level is required. Under such system-level data quality control, one of the basic requirements is that the data should follow the conservation of vehicles principle. However, this fundamental requirement and the associated diagnostic methods for identifying violations of this constraint have received little attention in the transportation engineering literature. This paper presents a methodology for checking conservation of vehicles over a series of detectors and adjusting the data using a constrained nonlinear optimization approach whenever the conservation principle is violated. The generalized reduced gradient method is adopted and applied to a 2-mi test bed in San Antonio. The method is validated using simulated data generated with CORSIM simulation software.

Transit passenger origin-destination estimation in congested transit networks with elastic line frequencies

This paper deals with the transit passenger origin-destination (O-D) estimation problem by using updated passenger counts in congested transit networks and outdated prior O-D matrix. A bilevel programming approach is extended for the transit passenger O-D updating problem where the upper-level problem seeks to minimize the sum of error measurements in passenger counts and O-D matrices, while the lower level is the stochastic user equilibrium assignment problem for congested transit networks. The transit assignment framework is based on a frequency-adaptive transit network model in this paper, which can help determine transit line frequencies and the network flow pattern simultaneously in congested transit networks. A heuristic solution algorithm is adapted for solving the transit passenger O-D estimation problem. Finally, a numerical example is used to illustrate the applications of the proposed model and solution algorithm.

Dynamic Origin–Destination Demand Estimation Using Automatic Vehicle Identification Data

This paper proposes a dynamic origin-destination (OD) estimation method to extract valuable point-to-point split-fraction information from automatic vehicle identification (AVI) counts without estimating market-penetration rates and identification rates of AVI tags. A nonlinear ordinary least-squares estimation model is presented to combine AVI counts, link counts, and historical demand information into a multiobjective optimization framework. A joint estimation formulation and a one-sided linear-penalty formulation are further developed to take into account possible identification and representativeness errors, and the resulting optimization problems are solved by using an iterative bilevel estimation procedure. Based on a synthetic data set, this study shows the effectiveness of the proposed estimation models under different market-penetration rates and identification rates