Disaggregate Travel Demand Models Research Articles

A Review of Activity-based Disaggregate Travel Demand Models

This paper reviews the literature on disaggregated travel demand models from a choice perspective, focusing on activity-based models (ABMs) and synthetic population generators as example approaches to capture these choice dimensions. We discuss the links between these models and elaborate on their integration, emphasizing how advanced population synthesis can address gaps in ABMs. The importance of incorporating multi-person and multi-period contexts to more accurately represent travel behavior is highlighted. Further advances aimed at improving the practical application of these models are also discussed.

A Complex Network Methodology for Travel Demand Model Evaluation and Validation

Travel demand can be viewed as a weighted and directed graph where nodes are the origins and destinations and links represent the trips between nodes. This paper presents a network-theoretic methodology to evaluate and validate travel demand models. We apply the proposed method on three disaggregate travel demand models from Melbourne, Australia. Statistical properties of the modeled networks are compared against the observed networks over time. The new approach reveals the network structure and connectivity of the modeled trips that are not usually captured by traditional evaluation and validation methods. Results demonstrate the complexity involved in the development, evaluation, and validation of travel demand models, which calls for advanced evaluation techniques reflecting a wide range of attributes of the observed and modeled data, travelers, mobility patterns, and complex network characteristics.

Travel demand forecasts improved by using cross-sectional data from multiple time points: enhancing their quality by linkage to gross domestic product

Forecasts using disaggregate travel demand models are often based on data from the most recent time point, even when cross-sectional data is available from multiple time points. However, this is not a good use of the data. In a previous study, the author proposed a method that jointly utilises cross-sectional data from multiple time points in which parameters are assumed to be functions of time (year), meaning that the parameter values vary over time. The method was applied to journey-to-work mode choice analyses for Nagoya, Japan. Behaviours in 2001 were forecast using one model with only the most recent 1991 dataset and other models that combined datasets for 1971, 1981, and 1991. The latter models outperformed the former model, which demonstrated the applicability of the proposed method. Although the functions of time ascribe the parameter changes to the trends over time, the theoretical underpinnings needed further investigation. The aim of this study is to analyse the same dataset used in the author’s previous study, but express the parameters as functions of gross domestic product (GDP) per capita. This method has fewer problems related to its theoretical underpinnings, since the parameter changes are explained by the effects of economic conditions. The functions of GDP per capita produced better forecasts than the functions of time. In addition, the functions of GDP per capita present fewer problems when choosing functional forms and extrapolating into the distant future/past and even to other areas. Sensitivity analysis with respect to uncertainties in the future GDP per capita showed that the proposed models are practical.

Travel demand forecasts improved by using cross-sectional data from multiple time points

Forecasts of travel demand are often based on data from the most recent time point, even when cross-sectional data is available from multiple time points. This is because forecasting models with similar contexts have higher transferability, and the context of the most recent time point is believed to be the most similar to the context of a future time point. In this paper, the author proposes a method for improving the forecasting performance of disaggregate travel demand models by utilising not only the most recent dataset but also an older dataset. The author assumes that the parameters are functions of time, which means that future parameter values can be forecast. These forecast parameters are then used for travel demand forecasting. This paper describes a case study of journeys to work mode choice analysis in Nagoya, Japan, using data collected in 1971, 1981, 1991, and 2001. Behaviours in 2001 are forecast using a model with only the most recent 1991 dataset and models that combine the 1971, 1981, and 1991 datasets. The models proposed by the author using data from three time points can provide better forecasts. This paper also discusses the functional forms for expressing parameter changes and questions the temporal transferability of not only alternative-specific constants but also level-of-service and socio-economic parameters.

Disaggregate Multimodal Travel Demand Modeling Based on Road Pricing and Access to Transit

In this paper, the conventional disaggregate travel demand model, a probability model for the modeling of multiple modes, generally called random utility maximization (RUM), is expanded to a model of count of mode choice. The extended travel demand model is derived from general economic theory—maximizing instantaneous utility on the time horizon, subject to a budget constraint—and can capture the dynamic behavior of countable travel demand. Because the model is for countable dependent variables, it has a more realistic set of assumptions to explain travel demand than the RUM model. An empirical test of the theoretical model using a toll facility user survey in the New York City area was performed. The results showed that the theoretical model explained more than 50% of the trip frequency behavior observed in New York City toll facility users. Travel demand for toll facility users increased with respect to household employment, household vehicle count, and employer payment of tolls and decreased in terms of travel time, road pricing, travel distance, and mass transit access.

A note on the consistent aggregation of nested logit demand functions

The present paper derives a set of rules allowing for the consistent aggregation of nested logit travel demand functions across origin and destination zones. Presented aggregation rules are derived for the case when the mode choice is performed conditional on destination choice. The derivation is based on the principles of consistency between aggregate and disaggregate travel demand models introduced by Sweet as well as upon the sampling theory.

Activity-based disaggregate travel demand model system with activity schedules

We present an integrated activity-based discrete choice model system of an individual’s activity and travel schedule, for forecasting urban passenger travel demand. A prototype demonstrates the system concept using a 1991 Boston travel survey and transportation system level of service data. The model system represents a person’s choice of activities and associated travel as an activity pattern overarching a set of tours. A tour is defined as the travel from home to one or more activity locations and back home again. The activity pattern consists of important decisions that provide overall structure for the day’s activities and travel. In the prototype the activity pattern includes (a) the primary – most important – activity of the day, with one alternative being to remain at home for all the day’s activities; (b) the type of tour for the primary activity, including the number, purpose and sequence of activity stops; and (c) the number and purpose of secondary – additional – tours. Tour models include the choice of time of day, destination and mode of travel, and are conditioned by the choice of activity pattern. The choice of activity pattern is influenced by the expected maximum utility derived from the available tour alternatives.

Travel demand model system for the information era

The emergence of new information technologies and recent advances in existing technologies have provided new dimensions for travel demand decisions. In this paper we propose a comprehensive travel demand modeling framework to identify and model the urban development decisions of firms and developers and the mobility, activity and travel decisions of individuals and households, and to develop a system of models that can be used by decision makers and planners to evaluate the effects of changes in the transportation system and development of information technologies (e.g. various tele-commuting, tele-services and Intelligent Transportation Systems). The implementation of an operational model system based on this framework is envisioned as an incremental process starting with the current “best practice” of disaggregate travel demand model systems. To this end, we present an activity-based model system as the first stage in the development of an operational model system.

A Model System for Worker's Daily Travel Behaviour Based on Disaggregate Model

Since middle 1970's, many studies have been carried out on disaggregate travel demand models. However, applications of those models to transportation planning are limited: they are usually applied to the local and short-term transportation planning, or they are used in the frame of traditional 4-step travel demand forecasting models by replacing some of those steps. That is, there have been few attempts to develop and apply disaggregate travel demand model systems for predicting travel demand in metropolitan areas.This paper develops a disaggregate model system for the travel demand of workers during a day in a metropolitan area.Daily travel pattern by an individual consists of visits to one or several destinations in trip chainnings that are movements away from his base (home or work place). We assume that these decisions during a day are not independent each other but there exist some relationships among trips that make up tours. Therefore, it is important to represent these relationships among trips in modeling the daily travel demand.In this study, by using the concept of utility maximization of base-to-base trip chainnig, we attempt to develop a set of models for predicting trip generation type, trip purpose, destination, mode choices in each trip on the way from work to home. The structure of travel demand model is represented by using nested logit model so that we avoid a simultaneous model's complexity caused by a large number of alternatives that a traveler faces in making his travel decision, while the model represents simultaneous decision-making process. The model incorporates the concept that decisions of a trip in a trip chain depend on decisions of trips conducted theretofore and decisions of trips planned thereafter, as well as on current trip conditions.The estimation of the model parameters was carried out by using a sequential approach. Empirical examinations of the model system for Nagoya metropolitan area show encouraging results.

A Model System for Predicting Urban Travel Demand Based on Disaggregate Behavioral Model -Non-worker Travel Demand Models-

Since middle 1970's many studies have been carried out on disaggregate travel demand models. However, applications of those models to transportation plannings are limited; they are usually applied to local and short-term transportation plannings, or they are used in the frames of traditional 4-step travel demand forecasting models by replacing some of those steps. That is, there have been few attempts to develop and apply disaggregate travel demand model systems for predicting travel demand in metropolitan areas.The model developed in this study incorporates trip frequency, destination and mode choice for non-woker travel by using the conception of travel tours into utility-maximizing frame. The model incorporates the concept that decisions of trips in a travel tour depend on decisions of trips conducted theretofore and decisions of trips planned thereafter, as well as on current trip conditions.Empirical examinations for Nagoya metropolitan area show the feasibility of the model.

Further evidence on the temporal stability of disaggregate travel demand models

Current evidence on the transferability of disaggregate travel demand models is inconclusive. Adding to this body of research, the present analysis focuses upon the temporal characteristics of work trip behavior in the San Francisco Bay Area. Using before and after data sets associated with the BART Impact Travel Study, multinomial logit models of work trip modal choice are estimated. The results indicate that the general form and the coefficient estimates of a pre BART model are transferable in time. Moreover, when updated to reflect BART's presence, the model's predictive success and its implied elasticity measures are generally accurate, relative to those implied by reestimating the entire model on post BART data. Finally, as economic theory would predict, elasticity measures of the service related variables were found to increase over time.

Behavioural theories of dispersion and the mis-specification of travel demand models

Conventional of first generation transport models have for some time been heavily criticised for their lack of behavioural content and inefficient use of data; more recently second generation or disaggregate travel demand models based on a theory of choice between discrete alternatives have also been viewed critically. First, it has been argued that implemented structures—and particularly the Multinomial Logit model—have not been sufficiently general to accommodate the “interaction” between alternatives; and second, and perhaps more importantly, that the underpinning theory, involving a perfectly discriminating rational man ( homo economicus), endowed with complete information is an unacceptable starting point for the analysis of behaviour. In this paper the potential errors in forecasting travel response arising from theoretical misrepresentation are investigated; more generally, the problems of inference and hypothesis testing in conjuction with cross-sectional models are noted. A framework is developed to examine the consequences of the divergence between the behaviour of individuals in a system, the observed, and that description of their behaviour (which is embedded in a forecasting model) imputed by an observer, the modeller. The extent of this divergence in the context of response to particular policy stimuli is examined using Monte Carlo simulation for the following examples: (i) alternative assumptions relating to the structure of models reflecting substitution between similar alternatives; (ii) alternative decision-making processes; (iii) limited information and “satisficing” behaviour; and (iv) existence of habit in choice modelling. The method has allowed particular conclusions to be nade about the importance of theoretical misrepresentation in the four examples. More generally, it highlights the problems of forecasting response with cross-sectional models and draws attention to the problem of validation which is all too often associated solely with the goodness of statistical fit of analytic functions to data patterns.

Modeling traveler responses to alternative gasoline allocation plans

This paper describes a modeling technique for estimating responses of urban travelers to gasoline shortages under three different procedures for allocating scarce gasoline supplies. The allocation procedures are: allow the price of gasoline to rise to a market clearing level, allocate gasoline by means of white-market coupons, and allocate gasoline by means of traditional rationing. The modeling technique is based on a system of disaggregate travel demand models and provides for representation of non-price restraints on gasoline consumption (e.g. traditional rationing) and for changes in multi-destination travel, as well as for changes in gasoline prices and in modes, destinations and frequencies of travel. The application of the modeling technique is illustrated by using it to estimate the effects of a gasoline shortage in the Washington, D.C., area. The results of this application indicate that in all allocation procedures tested, reductions in non-work trip frequencies and lengths are the main sources of gasoline savings in the presence of shortages. Transit improvements and carpool incentives moderate but do not prevent the effects of shortages on non-work travel. In price-based allocation methods, including allocation by means of white-market coupons, the reductions in travel that accompany a gasoline shortage are considerably larger among low-income households than they are among high-income ones. This distribution of effects is reversed when non-price-based allocation is used.

Reducing the aggregation bias: A heuristic approach

Disaggregate travel demand models are increasingly applied to predict aggregate demand. This is usually done by applying these models to zonal aggregated data. This prediction method causes a bias (“aggregation bias”) in the predictions obtained. Several aggregation procedures aimed at reducing this bias have been developed in recent years. Many of them, however, suffer from a number of drawbacks. First, no aggregation procedure can assure a priori the reduction of this bias to negligible proportions. Second, the direction of the residual bias is unknown. Third, no measure is provided for the residual aggregation bias. This paper presents a new approach to aggregate analysis. It is based on a heuristic model, which enables the application of diaggregate models to zonally aggregated systems for travel prediction. This heuristic decomposes the bias into two components: average systematic bias and a random variation around it. The systematic component is eliminated, such that the residual aggregation error is unbiased, its magnitude is small and a measure is provided to express this error quantitatively. The performance of the heuristic was demonstrated by an empirical study of mode choice in Washington, D.C. The heuristic was found efficient in reducing the bias and predicting the residual aggregation error.

Assumed versus estimated functional form in disaggregate mode choice models

The logit technique, commonly used to estimate disaggregate mode choice models, employs an assumed functional form which is restrictive and may produce biased estimates. As an alternative method of estimation, the extended cubic spline function which is found to have desirable properties, is proposed. In addition, it is shown for the bivariate case that restricted least squares which has been shown to be equivalent to cubic splines, provides an easier method for estimation. A disaggregate travel demand model is estimated using two techniques: the standard logit model and restricted least squares. The results indicate that the logit technique is highly susceptible to misspecification error since it fails to take account of locally dependent behavior. The restricted least squares method which estimates the functional form, is found to be more flexible, provide a richer menu of test statistics, and provide more reliable parameter estimates for public policy purposes.

A Disaggregate Travel Demand Model

"A Disaggregate Travel Demand Model." Journal of the Operational Research Society, 27(2), pp. 521–522

A Disaggregate Travel Demand Model

A Disaggregate Travel Demand Model

Aggregate and Disaggregate Travel Demand Models

An empirical investigation of the relative merits of aggregate and disaggregate travel demand models is carried out. Specifically, the empirical investigation concerns models of travel mode choice, using work-trip data for CBD trips in London, England, and Syracuse, N.Y. This research represents the first such comparison to be made between aggregate and disaggregate models of intraurban trips. Initially, the statistically best disaggregate models were constructed for each data set and estimation of the modal choices from the other data set was attempted. Statistical measures were computed for each model to assess the goodness-of-fit achieved. Subsequently, the data were aggregated to districts, based on the London Transportation Study, for the London data, and to zip code areas for the Syracuse data. Again, the statistically best aggregate models were constructed for each data set and the same statistical and estimation tests were made for the aggregate models.