Estimating origin-destination matrices from roadside survey data

The purpose of this study is to develop a valid and efficient method for estimating origin-destination tables from roadside survey data. Roadside surveys, whether conducted by interviews or postcard mailback methods, typically have in common the sampling of trip origin and destination information at survey stations. These survey stations are generally located where roads cross “screenlines,” which are imaginary barriers drawn to intercept the trip types of interest.Such surveys also include counts of traffic volumes, by which the partial origin-destination (O-D) tables obtained at the different stations can be expanded and combined to obtain the complete O-D table which represents travel throughout the entire study area. The procedure used to expand the sample O-D information from the survey stations must recognize and deal appropriately with a number of problems:

• 1.(i) The “double counting” problem: Long-distance trips may pass through more than one survey station location; thus certain trips have the possibility of being sampled and expanded more than once, leading to a potentially serious overrepresentation of long-distance trips in the complete expanded trip table.
• 2.(ii) The “leaky screenline” problem: Some route choices, particularly those using very lightly traveled roads, may miss the survey stations entirely, leading to an underestimation of certain O-D patterns, or to distorted estimates if such sites are arbitrarily coupled with actual nearby station locations.
• 3.(iii) The efficient use of the data: There is a need to adjust expansion factors to compensate for double counting and leaky screenlines. How can this be accomplished such that all of the data obtained at the stations are used without loss of information?
• 4.(iv) The consequences of uncertainty and unknown travel behavior: Since the O-D data and other sampled variables are subject to random error, and since in general the probability of encountering a long-distance trip at some survey stations is affected by traveler route-choice behavior, which is not understood, the sample expansion procedure must rely on the use of erroneous input data and questionable assumptions. The preferred procedure must minimize, rather than amplify, the effects of such input errors.

Here, five alternate methods for expanding roadside survey data in an unbiased manner are proposed and evaluated. In all cases, it is assumed that traveler route choice generally follows the pattern described by Dial's multipath assignment method. All methods are applied to a simple hypothetical network in order to examine their efficiency and error amplification properties. The evaluation of the five methods reveals that their performance properties vary considerably and that no single method is best in all circumstances. A microcomputer program has been provided as a tool to facilitate comparison among methods and to select the most appropriate expansion method for a particular application.     

Analysis of equilibrium automobile holdings in Israel with aggregate discrete choice models

Models of household vehicle ownership decisions do not suffice as a basis for forecasting the size and composition of aggregate vehicle holdings. Forecasting applications require that such models be imbedded in systems describing the operation of the automobile market. This paper presents a new model of short run equilibrium in the automobile market. The short run is a period within which new car designs and prices are fixed but used car prices adjust competitively to market forces. The magnitude and mix of new car sales, the extent of used car scrappage and the composition of used car holdings are determined in equilibrium with used car prices. An econometric version of the market model has been estimated on Israeli data and applied to analyze the impact of vehicle tax policy on automobile holdings in Israel. The paper describes this application.     

Estimating automobile utilisation with panel data: An investigation of alternative assumptions for the initial conditions and error covariances

This paper illustrates how recent developments in the econometrics of panel data analysis can be implemented in the study of the dynamics of continuous choice. Adopting the household's annual use of automobiles as the empirical setting with the data defined as a four-wave panel, we outline alternative assumptions for representing the role of unobserved heterogeneity, in particular the initial conditions, and the correlational structure between the observed and unobserved influences on choice. We show the importance of treating the initial conditions as endogenous. With the growing interest in integrating discrete and continuous choice models, typically through the application of the idea of selectivity, more thought has to be given to the error structure of the continuous choice module when the panel is short, as is so often the empirical reality. The simple application of econometric methods developed for long time series can be quite misleading.     

Estimating GEV models with censored data

We examine the problem of estimating parameters for Generalized Extreme Value (GEV) models when one or more alternatives are censored in the sample data, i.e., all decision makers who choose these censored alternatives are excluded from the sample; however, information about the censored alternatives is still available. This problem is common in marketing and revenue management applications, and is essentially an extreme form of choice-based sampling. We review estimators typically used with GEV models, describe why many of these estimators cannot be used for these censored samples, and present two approaches that can be used to estimate parameters associated with censored alternatives. We detail necessary conditions for the identification of parameters associated exclusively with the utility of censored alternatives. These conditions are derived for single-level nested logit, multi-level nested logit and cross-nested logit models. One of the more surprising results shows that alternative specific constants for multiple censored alternatives that belong to the same nest can still be separately identified in nested logit models. Empirical examples based on simulated datasets demonstrate the large-sample consistency of estimators and provide insights into data requirements needed to estimate these models for finite samples.     

Estimating vehicle miles traveled combined with ITS data

‘Vehicle miles traveled’ (VMT) is an important performance measure for highway systems. Currently, VMT [or ‘annual average daily traffic’ (AADT)] is estimated from a combination of permanent counting stations and short-term counts done at specified locations as part of the Highway Performance Monitoring System (HPMS) mandated by the US Federal Highway Administration. However, on some roadway sections, Intelligent Transportation Systems (ITS) such as detectors and cameras also produce traffic data. The question addressed in this paper is whether and under what conditions ITS systems data could be used instead of HPMS short-term counts (called ‘coverage counts’)? This paper develops a methodology for determining a threshold number of missing daily traffic counts, or alternatively, the number of valid ITS data observations needed, in order to confidently replace the HPMS coverage counts with ITS data.

Because ITS counts, coverage counts, and actual ground counts (e.g. continuous counts) cannot be found coexisting on a roadway section, it is hard to compare them directly. In this paper, the Monte Carlo simulation method is employed to generate synthetic ITS counts and coverage counts from a set of relatively complete traffic counts collected at a continuous count station. Comparisons are made between simulated ITS counts, coverage counts, and actual ground counts. The simulation results indicate that when there are<330 daily traffic counts missing in a set of ITS counts in a year, that is, when there are at least 35 days of valid data, ITS counts can be used to derive a better AADT than using coverage counts. This result is applied to calculate the VMT for the Hampton Roads region in Virginia. The comparison between the VMTs derived with using and not using the threshold number indicates that these two VMTs are significantly different.     

A framework for analyzing rank-ordered data with application to automobile demand

In this paper, we develop a general random utility framework for analyzing data on individuals’ rank-orderings. Specifically, we show that in the case with three alternatives one can express the probability of a particular rank-ordering as a simple function of first choice probabilities. This framework is applied to specify and estimate models of household demand for conventional gasoline cars and alternative fuel vehicles in Shanghai based on rank-ordered data obtained from a stated preference survey. Subsequently, the framework is extended to allow for random effects in the utility specification to allow for intrapersonal correlation in tastes across stated preference questions. The preferred model is then used to calculate demand probabilities and elasticities and the distribution of willingness-to-pay for alternative fuel vehicles.     

Discrete choice with spatial correlation: A spatial autoregressive binary probit model with endogenous weight matrix (SARBP-EWM)

Discrete choice modeling is widely applied in transportation studies. However, the need to consider correlation between observations creates a challenge. In spatial econometrics, a spatial lag term with a pre-defined weight matrix is often used to capture such a correlation. In most previous studies, the weight matrix is assumed to be exogenous. However, this assumption is invalid in many cases, leading to biased and inconsistent parameter estimates. Although some attempts have been made to address the endogenous weight matrix issue, none has focused on discrete choice modeling. This paper fills an existing gap by developing a Spatial Autoregressive Binary Probit Model with Endogenous Weight Matrix (SARBP-EWM). The SARBP-EWM model explicitly considers the endogeneity by using two equations whose error terms are correlated. Markov Chain Monte Carlo (MCMC) method is used to estimate the model. Model validation with simulated data shows that the model parameters can converge to their true values and the endogenous weight matrix can be reliably recovered. The model is then applied to a simplified firm relocation choice problem, assuming that similar size firms influence one another. The model quantifies the peer effect, and takes into consideration other independent variables including industry type and population density. The estimation results suggest that peer influence among firms indeed affect their relocation choices. The application results offer important insights into business location choice and can inform future policy making. The sample size for applying the model is currently limited to hundreds of observations. This paper contributes to the existing literature on discrete choice modeling and spatial econometrics. It provides a new tool to discover spatial correlations that are hidden in a wide range of transportation issues, such as land development, location choice, and various travel behavior. Those hidden spatial correlations are otherwise difficult to identify and estimation results may be biased. Establishing a new model that explicitly considers endogenous weight matrix and applying the model to a real life transportation issue represent a significant contribution to the body of literature.     

Preference heterogeneity in mode choice based on a nationwide survey with a focus on urban rail

The provision of efficient and effective urban public transport and transport policy requires a deep understanding of the factors influencing urban travellers’ choice of travel mode. The majority of existing literature reports on the results from single cities. This study presents the results of a nationwide travel survey implemented to examine multiple modes of urban passenger transport across five mainland state capitals in Australia, with a focus of urban rail. The study aims to explore differences in mode choices among surveyed travellers sampled from the five cities by accounting for two types of factors: service quality and features of public transport, and socio demographic characteristics. A stated preference approach is adopted to elicit people’s valuation of specified mode-choice related factors and their willingness to pay. In particular, the availabilities of wireless and laptop stations – two factors rarely examined in the literature, were also considered in the SP survey. The survey data were analysed using mixed logit models. To test for preference heterogeneity, socio-demographic factors were interacted with random parameters, and their influences on marginal utilities simulated. The analysis reveals that intercity differences, user group status, gender, income, and trip purposes partially explain observed preference heterogeneity.     

Non-trading,lexicographic and inconsistent behaviour in stated choice data

This paper discusses a number of issues relating to the pre-analysis and cleaning of stated choice data, where we look specifically at the problems caused by non-trading, lexicographic and inconsistent response patterns. We argue that this process is considerably more complex and challenging than many in the field have hitherto acknowledged, with the standard practice being the use of rather ad hoc procedures for the identification of these phenomena. A detailed analysis on four different stated choice datasets highlights the potential impacts of these methods on model estimation results.     

Web-based survey design for unravelling semi-compensatory choice in transport and urban planning

The estimation of semi-compensatory models is gaining momentum in transport planning in recent years. However, traditional survey methodologies focus on collecting solely compensatory choice data, which leads to information loss when semi-compensatory models are estimated. The present study proposes a novel web-based survey that enables collecting data about the entire semi-compensatory choice process. The web-based environment allows seamless tracking of semi-compensatory choice protocols without interfering with the natural choice process and without introducing problems related to comprehension bias, narrative inconsistency and misinterpretation of the choice protocols. The procedure is applied to rental apartment choice by students and results shed light on semi-compensatory choice by: (1) demonstrating the importance of choice set formation; (2) unravelling the distribution of threshold selection across the population; (3) revealing the linkage between the viable choice-set and the choice.     

Understanding suburban travel demand: Flexible modelling with revealed and stated choice data

We analyse the choice of mode in suburban corridors using nested logit specifications with revealed and stated preference data. The latter were obtained from a choice experiment between car and bus, which allowed for interactions among the main policy variables: travel cost, travel time and frequency. The experiment also included parking cost and comfort attributes. The attribute levels in the experiment were adapted to travellers’ experience using their revealed preference information. Different model specifications were tested accounting for the presence of income effect, systematic taste variation, and incorporating the effect of latent variables. We also derived willingness-to-pay measures, such as the subjective value of time, that vary among individuals as well as elasticity values. Finally, we analysed the demand response to various policy scenarios that favour public transport use by considering improvements in level-of-service, fare reductions and/or increases in parking costs. In general, demand was shown to be more sensitive to policies that penalise the private car than those improving public transport.     

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.     

Consumer preferences for battery electric vehicles: A choice experimental survey in China

To accelerate the diffusion of battery electric vehicles (BEVs), consumer preferences for different products and policy attributes must be determined. Although previous studies have investigated consumer preferences for some product attributes, including purchase price, operation cost, driving range, and charging time, limited studies have discussed the broader aspects of product attributes, such as battery warranty and depreciation rate. Moreover, market-oriented incentives, including the personal carbon trading (PCT) scheme and the tradable driving credits (TDC) scheme, can theoretically be effective alternatives to expensive purchase subsidies. However, there is a lack of empirical evidence that confirms the influence of these two schemes on BEV adoption. To fill these gaps, we conducted a stated preference choice experimental survey in China and investigated the effect of product attributes, existing policy incentives, and two emerging market-oriented incentives on BEV adoption. Our results reveal that along with the main product attributes, battery warranty has a significant positive effect on inducing mainstream consumers to adopt BEVs while no preference difference occurs among existing policy incentives after purchase subsidies are abolished. For young consumers, almost all incentives that reduce the operation cost (e.g., PCT) or increase convenience (e.g., TDC) can increase their adoption of BEVs. These findings can provide important implications for the government with regard to designing novel incentives and promoting BEV adoption.     

Random regret minimization or random utility maximization: an exploratory analysis in the context of automobile fuel choice

Interest in alternative behavioural paradigms to random utility maximization (RUM) has existed ever since the dominance of the RUM formulation. One alternative is known as random regret minimization (RRM), which suggests that when choosing between alternatives, decision makers aim to minimize anticipated regret. Although the idea of regret is not new, its incorporation into the same discrete choice framework of RUM is very recent. This paper is the first to apply the RRM‐model framework to model choice amongst durable goods. Specifically, we estimate and compare the RRM and RUM models in a stated choice context of choosing amongst vehicles fuelled with petrol, diesel and hybrid (associated with specific levels of fuel efficiency and engine capacity). The RRM model is found to achieve a marginally better fit (using a non‐nested test of differences) than its equally parsimonious RUM counterpart. As a second contribution, we derive a formulation for regret‐based elasticities and compare utility‐based and regret‐based elasticities in the context of stated vehicle type choices. We find that in the context of our choice data, mean estimates of elasticities are different for many of the attributes and alternatives. Copyright © 2011 John Wiley & Sons, Ltd.     

Influence of the weather on mode choice in corridors with time-varying congestion: a mixed data study

Studies that link human behaviour to the influence of weather have historically been conducted in such fields as tourism, marketing and leisure. In most studies that jointly examine weather and the mode of transport, only open-air transportation has been considered (for example, bicycle, motorcycle or walking). This focus, together with the habitual use of data collected with automatic devices and a lack of studies that analyse this issue using stated preference data, are the main reasons motivating this paper. This paper aims to analyse the influence of weather and the density of traffic on the choice of transport mode. A case study is conducted in an access/egress corridor located in the city of Barcelona (Spain). Two data sources were used: revealed preference and stated preference data. Modelling techniques using mixed data enabled the stronger features from both data sources to be captured. Finally, we discuss how the selection of different alternative specific constants in models estimated using mixed data could generate unrealistic forecasting results if environmental changes are expected in the actual market.     

The sensitivity of parameter estimates to data specification in mode choice models

Standard network data are generally used in estimation of mode choice models. These data are inaccurate in several ways, but the cost of correcting the inaccuracies is great. This paper analyzes the effects which correcting some of the inaccuracies in the standard network data has on the estimated parameters of mode choice models. Models are estimated on the standard network data and on data which have been adjusted so as to correct the problems in the standard network data. It is found that, for analysis of policies affecting transfer wait times or distances to bus stops, correction of the standard network data is advisable. For other policy analyses, however, it seems that the extra expense of correcting the standard data is unnecessary.     

A dichotomous choice survey for quantifying option and non-use values of bus services in Korea

The total economic value for a transportation service consists of use, option, and non-use value. The use benefit is based on a traveler’s willingness to pay for usual consumption of the service. The optional value, on the other hand, is related to the possible use of the service for trips not yet anticipated or currently accommodated by other travel modes. The non-use value, however, is derived from the intrinsic merit of the service, even though a trip-maker never actually or potentially depends on the mode. A closed-ended contingent valuation method is considered for the quantification of the option and non-use values. A survey of single- and double-bounded dichotomous choices is conducted with a case study of South Korean bus operations. A logistic regression model and a survival analysis for the single- and double-bounded approaches, respectively, are applied. The estimation result is examined according to the statistical property required and the behavioral validity expected. In particular, three issues from the output are discussed. First, the results help to show the preferable framework between single- and double-bounded surveys for addressing an individual’s option and non-use values. Second, the differences in the absolute values of option and non-use values are compared. Thirdly, the relationship between trip-makers’ willingness to pay and the level of service of their primary travel modes are investigated. In conclusion, the summary of research and the possibilities for future studies are given.     

A long panel survey to elicit variation in preferences and attitudes in the choice of electric vehicles

This paper describes the methodology we set up to gather appropriate data to study the impact of real life experience with electric vehicles (EVs) over a relatively long period of time on individual preferences and attitudes. We used stated choices (SC) to elicit individual preferences because EVs and their associated charging infrastructure are not yet fully integrated onto the market. Furthermore, to measure the extent to which the experience of using an EV may affect individual preferences and attitudes, we set up a “long panel” survey, where data was gathered before and after individuals experienced an EV in real life during a three-month period. We also measured attitudinal effects (AE) that might affect the choice of an EV by individuals. To our knowledge, this represents the first example of a “long panel” SC/AE and the first attempt to measure the formation of preferences and attitudes for this emerging product. Our results show that preferences and attitudes are indeed affected by real life experience. In the SC experiment, the respondents only chose the EV half as often as compared to the situation where they had not yet tried it. Furthermore, we measured a change in attitude for statements regarding the use of EVs. On the whole, respondents got a more positive view of the EV driving performance and this change is significantly greater for women than for men. However, respondents expressed more concern about being able to maintain current mobility with an EV. The data gathered in this survey should also serve to analyse the changes generated by direct experience with EVs, and eventually to formulate and estimate advanced discrete choice models that allow insights into factors relevant for improved understanding of market behaviour.     

Importance of scenic byways in route choice: a survey of driving tourists in the United States

In order to develop more effective tourist information systems for use along scenic byways, it is important to know the characteristics of those people who include the presence of scenic byways in their selections of routes. Gaining a better understanding of these characteristics was the purpose of this study. The data presented here are from a United States (US) survey of the driving tourist's information needs and preferences weighted to be representative of US tourist travelers. The study showed that when planning a route to a destination on an overnight automobile trip, the driving tourist is most concerned with factors related to the actual driving of the route, such as the directness, safety, amount of congestion, and distance. Of secondary importance are factors that make the route entertaining or pleasant to drive, including whether the route is a scenic byway. Analysis of the importance of scenic byways by several demographic factors showed little difference in importance ratings except for age and household income. Examination of importance ratings by trip characteristics showed that the presence of scenic byways in selecting a route was more important for the traveler whose trip purpose is a vacation, who is in the midst of a long distance and duration trip, who will be either camping or staying in a hotel, and who has planned the trip well in advance. These results suggest that scenic byways are an excellent area for the implementation and testing of in-vehicle information systems for the driving tourist.     

Modelling commuting mode choice with explicit consideration of carpool in the choice set formation

This article investigates the carpool mode choice option in the context of overall commuting mode choice preferences. The article uses a hybrid discrete choice modelling technique to jointly model the consideration of carpooling in the choice set formation as well as commuting mode choice together with the response bias corrections through the accommodation of measurement equations. A cross-nested error structure for the econometric formulation is used to capture correlations among various commuting modes and carpool consideration in the choice set. Empirical models are estimated using a data set collected through a week-long commuter survey in Edmonton, Alberta. The empirical model reveals many behavioural details of commuting mode choice and carpooling. Interestingly, it reveals that interactions between various Travel Demand Management (TDM) tools with the carpooling option can be different at different level of decision making (choice set formation level and final choice making level).