Space-time dynamics: A modeling approach for commuting departure time on linked datasets

Commuters' departure time related decisions are important in time geography. Analytic tools have been proposed to capture the inherent choice determinants both in time and space. Although the dynamic aspects of the problem have been identified, most of the existing studies are based on static models. In this paper, a dynamic modeling framework is proposed to explore the relationship between commuters' departure time choices and the evolution of en route traffic. A data linkage method is developed to create an integrated dataset that enables the observation of commuters' reaction to changes in travel time and traffic conditions over time. A regional household travel survey is linked to travel information obtained from the Google Maps application program interface (API), creating a synthetic longitudinal dataset. Two decision rules are applied to model commuters' response to the evolution of traffic. The results indicate that travel time, distance to work location, flexibility in working schedule, expected arrival time, and commuters' sociodemographic influence departure time choices. It is also found that accounting for dynamics improves model fit and out-of-sample predictions. Both the dynamic model and the proposed data linkage method contribute to the understanding of human activities in space and time and can be used to enhance transportation demand analysis and urban policy studies.     

Dynamic analysis of air travel demand in competitive island markets

This paper describes a dynamic demand model, referred to as dynamic abstract mode model, for estimating both the short- and long-term responses of air passengers to changes in relative air-sea travel cost components in competitive markets. The implementation of the model in the competitive market of Aegean islands in Greece demonstrates the importance of considering the past volumes of air passengers and relative travel cost components to explain current air travel demand.     

A dynamic demand analysis of the United States air-passenger service

This study examines the short- and long-run effects of various determinants on the demand for US air passenger-services using the Johansen cointegration analysis and a vector error-correction (VEC) model. Results show that, in the long-run, airfare, disposable income and NASDAQ have significant effects on US air travel demand. The combined short-run dynamic effects of disposable income, NASDAQ, population and airfare jointly explain changes in air passenger-miles. Finally, we find that the 9/11 terrorist attacks drop air passenger demand by 5% during 2001:Q3-2002:Q2, which in turn pushes down the seat capacity by 4%. However, it has little impact on airfare.     

A cointegration analysis of bilateral air travel flows: The case of international travel to and from the United States

Demand for international air travel has risen over the past decade causing international visitation to the US to reach a record high in 2012. This paper assesses the dynamic impacts of GDP, exchange rate, and the 9/11 terrorist attacks on bilateral air travel flows between the US and its 11 major travel and trading partners. An autoregressive distributed lag modeling approach is employed to estimate short- and long-run relationships between variables. Long-run results demonstrate foreign GDP as the major determinant of demand for inbound travel to the US and US GDP is a crucial factor affecting demand for outbound travel from the US. These findings support a strong linkage between economic growth and demand for international air travel. The real exchange rate has relatively little impact on the bilateral air travel flows. The US dollar appreciation against foreign currencies is found to reduce demand for inbound travel to the US, while having mixed effects on outbound travel from the US. In the short-run, economic growth tends to be a primary factor influencing international travel flows to and from the US. The 9/11 market shock has a detrimental short- and long-run effect on the bilateral air travel flows, implying that the impact of 9/11 is prolonged in international air travel markets.     

Errors in variables in multinomial choice modeling: A simulation study applied to a multinomial logit model of travel mode choice

Modeling travel demand is a vital part of transportation planning and management. Level of service (LOS) attributes representing the performance of transportation system and characteristics of travelers including their households are major factors determining the travel demand. Information on actual choice and characteristics of travelers is obtained from a travel survey at an individual level. Since accurate measurement of LOS attributes such as travel time and cost components for different travel modes at an individual level is critical, they are normally obtained from network models. The network-based LOS attributes introduce measurement errors to individual trips thereby causing errors in variables problem in a disaggregate model of travel demand. This paper investigates the possible structure and magnitude of biases introduced to the coefficients of a multinomial logit model of travel mode choice due to random measurement errors in two variables, namely, access/egress time for public transport and walking and cycling distance to work. A model was set up that satisfies the standard assumptions of a multinomial logit model. This model was estimated on a data set from a travel survey on the assumption of correctly measured variables. Subsequently random measurement errors were introduced and the mean values of the parameters from 200 estimations were presented and compared with the original estimates. The key finding in this paper is that errors in variables result in biased parameter estimates of a multinomial logit model and consequently leading to poor policy decisions if the models having biased parameters are applied in policy and planning purposes. In addition, the paper discusses some potential remedial measures and identifies research topics that deserve a detailed investigation to overcome the problem. The paper therefore significantly contributes to bridge the gap between theory and practice in transport.     

Modelling dependence between tourism demand and exchange rate using the copula-based GARCH model

This paper investigates dependence between tourism demand and exchange rate, using the case of China, and from a new perspective by using copula–GARCH models. The empirical results show that the volatility of exchange rate is not a determinant factor in fluctuation of China's inbound tourism demand from the countries being studied. Furthermore, only Russia exhibits risk-adverse behaviour with extreme SUR depreciation, or CNY appreciation associated with an extreme decline in arrivals. Third, introducing the tail dependence and dynamic dependence between growth rates of tourism demand and exchange rate add much to the explanatory ability of the model. The findings of this study have important implications for destination manager and travel agent as it helps to understand the impact of exchange rates on China inbound tourism demand and provide a complementary academic approach on evaluating the role of exchange rates in the international tourism demand model.     

考虑换乘费用的城市公共交通票价优化

换乘所产生的附加费用,如时间和票价等,导致配流影响因素产生变化,如果沿用传统方式将降低预测精度和可靠性。通过引入换乘次数和方式等因子计算出行等待、乘车、换乘及风险评估预留时间等,定义广义出行费用与计算方法;建立双层规划模型求解最优票价,最后通过算例分析弹性出行需求、换乘费用、票价之间的关系。计算结果表明,换乘费用对出行需求的影响小于票价优化对出行需求的影响,优化票价随换乘费用增加而加速降低,为公共交通票价优化提供研究依据。     

Anticipating long-distance travel shifts due to self-driving vehicles

Increasing population and travel demand has prompted new efforts to model travel demand across the United States. One such model is rJourney that estimates travel demand among thousands of regions and models mode and destination choice. rJourney includes records representing 1.17 billion long-distance trips throughout the year 2010. Although inter-regional impacts caused by an increase of automated vehicles (AVs) has been investigated, there is little research on inter-regional travel and how longer distance destination and mode choices will change. Because of conveniences offered by AVs, the value of travel time of drivers is expected to fall, thus reducing the generalized cost of AV travel. To initially analyze the impacts of AVs in the United States, a new AV mode was added to a subset of the rJourney mode and destination choice models. With an initial scenario assuming an operating cost of AVs that is 118% of traditional cars, two outcomes are observed that are solely based on model results. First, the attractiveness of AVs severely digs into the airline travel market, reducing airline revenues to 53%. Second, the introduction of AVs results in a shift of destination choice, increasing travel in further distances for personal vehicles, but favoring closer distances across all modes, for an overall 6.7% decline in US passenger-miles traveled on existing long-distance trips. While this preliminary research has revealed an initial perspective on how an existing model can support AVs, the increasing availability of data as AVs emerge will refine nationwide long-distance modeling.     

Analysing the spatial-temporal characteristics of bus travel demand using the heat map

As the basic travel service for urban transit, bus services carry the majority of urban passengers. The characterisation of urban residents' transit trips can provide a first-hand reference for the evaluation, management and planning of public transport. Over the past two decades, data from smart cards have become a new source of travel survey data, providing more comprehensive spatial-temporal information about urban public transport trips. In this paper, a multi-step methodology for mining smart card data is developed to analyse the spatial-temporal characteristics of bus travel demand. Using the bus network in Guangzhou, China, as a case study, a smart card dataset is first processed to quantitatively estimate the travel demand at the bus stop level. The term ‘bus service coverage’ is introduced to map the bus travel demand from bus stops to regions. This dataset is used to create heat maps that visualise the regional distribution of bus travel demand. To identify the distribution patterns of bus travel demand, two-dimensional principal component analysis and principal component analysis are applied to extract the features of the heat maps, and the Gaussian mixture model is used for the feature clustering. The proposed methodology visually reveals the spatial-temporal patterns of bus travel demand and provides a practical set of visual analytics for transit trip characterisation.     

The built environment and vehicle ownership modeling: Evidence from 32 diverse regions in the U.S.

Vehicle ownership is an important determinant of the travel demand forecasting process. Vehicle ownership models are used by policy makers to identify factors that affect vehicle miles traveled, and therefore address problems related to energy consumption, air pollution, and traffic congestion. For the conventional travel demand forecasting, it logically follows land use forecasting, before trip generation, which is commonly treated as step one. The most critical limitation of the vehicle ownership models, especially in the conventional process, is that they are often related mainly to sociodemographic variables, not so much to built environmental variables. In this study, by pooling regional household travel survey data from 32 diverse regions (almost 92,000 households) of the U.S., and by controlling for socio-demographic and the built environmental variables, we estimated a vehicle ownership model that contributes to the understanding of vehicle ownership and improves the accuracy of travel demand forecasts. Two main findings of this research are: 1) The number of vehicles owned by a household increases with socio-demographic variables and decreases with almost all of the built environmental variables. For the urban planning and design practices, this finding suggests that car shedding occurs as built environments become more dense, mixed, connected, and transit-served. 2) We used both count regression and discrete choice models, and the results suggest that count regression models have better predictive accuracy. The model developed in this study can be directly used for travel demand modeling and forecasting by metropolitan planning organizations.     

Time-use data, analysis and modeling: toward the next generation of transportation planning methodologies

This paper argues that transportation planning methodologies must be built on the central thesis of the activity-based approach to travel demand modeling, namely, that travel is a derived demand that reflects people's desire and need to participate in activities. The paper discusses why this foundation for transportation planning methodologies is necessary to address contemporary planning and policy analysis issues. The paper also argues that the introduction of time-use data, analysis and modeling is a key element in the development of the next generation of transportation planning methodologies. Following a brief review of time-use studies, the paper discusses a number of planning and policy analysis areas in which time-use data will be of particular value, including the evaluation of induced or suppressed travel demand. The concepts advanced in the paper are illustrated with two brief numerical examples. These examples show how model systems based on time-use data can be used to (i) estimate the number of induced trips that would result from a reduction in commute travel time, and (ii) evaluate the impacts of alternative transportation improvement projects.     

Evolution of a dynamic ridesharing system based on rational behavior of users

A dynamic ridesharing system (DRS) is a system where users can find ridesharing partner(s) at any time, even shortly before making a trip. A DRS that does not consider individual preferences may cause dissatisfied matchings of users in a shared vehicle and lead to abandonment of DRS in the long term. To investigate the evolution of DRS, such as long-term adoption, this study develops a model of DRS considering the rational behavior and learning process of its users. User behavior is considered as travel mode choice and ridesharing partner choice decisions under the expected utility maximization concept. The day-to-day evolution of a DRS is simulated based on the proposed model, and the effects of user learning behaviors and some social factors pertinent to long-term DRS adoption are investigated.     

Representing pedestrian activity in travel demand models: Framework and application

There have long been calls for better pedestrian planning tools within travel demand models, as they have been slow to incorporate the large body of research connecting the built environment and walking behaviors. Most regional travel demand forecasting performed in practice in the US uses four-step travel demand models, despite advances in the development and implementation of activity-based travel demand models. This paper introduces a framework that facilitates the abilities of four-step regional travel models to better represent walking activity, allowing metropolitan planning organizations (MPOs) to implement these advances with minimal changes to existing modeling systems. Specifically, the framework first changes the spatial unit from transportation analysis zones (TAZs) to a finer-grained geography better suited to modeling pedestrian trips. The MPO's existing trip generation models are applied at this spatial unit for all trips. Then, a walk mode choice model is used to identify the subset of all trips made by walking. Trips by other modes are aggregated to the TAZ level and proceed through the remaining steps in the MPO's four-step model. The walk trips are distributed to destinations using a choice modeling approach, thus identifying pedestrian trip origins and destinations. In this paper, a proof-of-concept application is included to demonstrate the framework in successful operation using data from the Portland, Oregon, region. Opportunities for future work include more research on the potential routes between origins and destinations for walk trips, application of the framework in another region, and developing ways the research could be implemented in activity-based modeling systems.     

Joint estimation of mode and time of day choice accounting for arrival time flexibility,travel time reliability and crowding on public transport

This study develops joint choice models of mode and departure time for implementation in MetroScan, a new version of TRESIS (Hensher and Ton, 2002). Separate models are estimated for work and non-work purposes, testing all practical alternatives of model structure with a rich set of explanatory variables. The contributions of the current work to the existing TRESIS are twofold. First, travel demand for non-work purposes such as shopping, social and recreation are explicitly modelled in MetroScan as opposed to TRESIS that scales the demand for work purposes to obtain non-work travel demand. Second, choices of travel mode and departure time become more sensitive to situational factors such as the flexibility of arrival time, the reliability of travel time and crowding. Willingness to pay for various improvements to the level of service is derived. We describe and demonstrate how the proposed models are applied in the general modelling framework of MetroScan.     

Mitigating the impact of errors in travel time reporting on mode choice modelling

Travel time is a major component in understanding travel demand. However, the quantification of demand and forecasting hinges on understanding how travel time is perceived and reported. Travel time reporting is typically subject to errors and this paper focuses on the mitigation of their impact on choice models. The aim is to explain the origin of these errors by including elements of travel behaviour (e.g., activities during the trip), which have been shown to significantly affect mode choices and commuting satisfaction. Based on responses from a revealed preferences survey, we estimate a mode choice model that treats travel time as a latent variable and incorporates different sources of data along with information on travel activities. Employing these multiple – sometimes incongruent – sources of information in the choice model appears to be beneficial. Results from comparing a logit model assuming error-free inputs and the integrated hybrid model revealed significant impacts on the generated policy scenarios. The model results also contributed to identifying the main travel activity features that affect travel time reporting, providing indications that can assist in understanding and mitigating the impact of imprecise measurements.     

Investigating household vehicle ownership,mode choice and trip sharing decisions using a combined revealed preference/stated preference Nested Logit model: case study in Bangkok Metropolitan Region

This study proposes a Nested Logit model to investigate household travel behaviour in respect to vehicle ownership, mode choice and trip sharing decisions. The model is analysed using revealed preference (RP) and stated preference (SP) data since a combined estimation of RP/SP data is an effective method of expressing complex travel behaviour and forecasting travel demand for new transport services. In the proposed model, the nesting structure has two levels. The upper level shows car ownership, motorcycle ownership, and no vehicle-ownership choices, and the lower level shows the mode choice combinations for two-traveller households. Trip sharing is considered as one of the mode-choice options in the model. The proposed model is analysed using data from the Bangkok Metropolitan Region. The analysis conducted informs that Central Business District (CBD) travel, long distance travel, household income, job status, age of travellers and presence of school children in households are key aspects in household travel decisions. Based on these aspects, households make important decisions on vehicle ownership, mode choice and trip sharing. In addition, this study reveals commuters’ hidden preferences for modes that are not in existence, in particular the Mass Rapid Transit System in the Bangkok Metropolitan Region due to be fully implemented in 2010.     

The dynamics of modal split for freight transport

The paper presents a dynamic model of modal split in a multimodal freight transport system, which supposes that the evolution over time of transport demand is accompanied by a corresponding evolution of transport modes, and that users react with delay to cost variations. Starting with these hypotheses, and following the paradigm of random utility, a recursive equation is obtained, whose iterated application furnishes the sequence of the demand fractions on the various transport modes in the successive epochs of the time period during which the evolution of the transport system is studied and enables forecasting the future modal split evolution.     

甘肃省高速铁路对城市可达性及经济联系格局影响研究

高速铁路的开通会产生时空压缩效应,使城市之间的平均旅行时间减少,提升城市间经济联系度。借助可达性系数和引力模型,探讨高速铁路开通前后甘肃段城市可达性和经济联系的演变。高速铁路开通后,加权平均旅行时间缩短,甘肃省城市可达性得到提升,形成以兰州-天水-定西为交通中心,沿高速铁路为轴线对外辐射的空间格局趋势;城市之间的经济联系强度大幅度增强,城市间经济联系差异扩大;内部经济联系极化效应加剧,非高速铁路区存在形成交通和经济双重洼地的风险,可达性和经济联系度呈现部分空间耦合。     

Predicting the distribution of households and employment: a seemingly unrelated regression model with two spatial processes

Household and employment counts (by type) are key inputs to models of travel demand and air quality. For a variety of reasons, spatial dependence is very likely present in and across these counts. In order to identify the nature of these unobserved relationships, this study provides the first application of a feasible generalized spatial 3SLS estimation procedure for a seemingly unrelated regression (SUR) model with two spatial processes. Statistical tests reveal that this more generalized model is superior to its constrained versions (e.g., SUR models without spatial components or with just a spatial lag or spatial error process).In the resulting model of Austin, Texas data, local land-use conditions offer substantial predictive power of household and job densities, and transportation access plays a role, as anticipated. The work demonstrates that SUR estimation of land-use intensities from parcel-level data with two types of spatial dependence is feasible and meaningful. Coupled with an upstream model of land-use type, this work offers key inputs for travel demand analyses, with transportation system performance feedback.     

Level of aggregation of zoning and temporal transferability of the gravity distribution model: The case of Lyon

This study deals with the temporal transferability of the parameters of the gravity model of trip distribution and focuses on the trade-off between spatial resolution and data requirements. The models are calibrated using O–D matrices constructed from the three most recent Lyon household travel surveys (1985, 1995 and 2006) and generalised travel time data from coded transport networks for the three dates. Calibration has been conducted for three different zoning levels which have been chosen in line with common practice. The parameters obtained from model calibration are then applied to estimate O–D matrices at a later date and the results are compared using indicators that have been established for the zoning level applied in calibration, but also using indicators that have been aggregated in two different ways: aggregation to create larger zones or distance segments.Our findings confirm our initial intuition: the choice of zoning is fundamentally important. Moreover, in the best case, the parameters of the model change, but not sufficiently for the goodness-of-fit of the “predicted” model to be very different from that of the matrix obtained during calibration. It is possible to use the gravity model for forecasting purposes, but on condition that the goals of the study are compatible with the level of error in the reproduction of the observed matrices. If the zoning is either too coarse or too fine grained, forecasting performance is compromised.