Travel Demand Model Research Articles

Empirical Modeling Analysis of Potential Commute Demand for Carsharing in Shanghai, China

Carsharing is an emerging commute mode in China, which may produce social and environmental benefits. This paper aims to develop a commute mode choice model to explore influential factors and quantify their impacts on the potential demand for carsharing in Shanghai. The sample data were obtained from a revealed preference (RP) and stated preference (SP) survey and integrated with level-of-service attributes from road and transit networks. The RP survey collected commuters’ trip information and socioeconomic and demographic characteristics. In the SP survey, four hypothetical scenarios were designed based on carsharing’s unit price to collect commuters’ willingness to shift to carsharing. Data fusion method was applied to fuse RP and SP models. The joint model identified the target group of choosing carsharing with certain socioeconomic and demographic attributes, such as gender, age, income, household member, household vehicle ownership, and so on. It also indicates that the value of time (VOT) for carsharing is 35.56 RMB Yuan (5.08 US Dollar)/h. The elasticity and marginal effect analysis show that the direct elasticity of carsharing’s fare on its potential demand is −0.660, while the commuters, who have a more urgent plan on car purchase or are more familiar with the carsharing service, have much higher probabilities to choose carsharing as their commute modes. The developed model is expected to be applied to the urban travel demand model, providing references for the formulation of carsharing operation scheme and government policy.

An Empirical Activity Sequence Approach for Travel Behavior Analysis in Vilnius City

The approach defines the process of conducting an empirical research of the travel behavior patterns of residents of Vilnius city. It defines survey methodology and important mobility parameters such as activity sequences and their probabilities of homogeneous urban population segments during the weekday. This empirical research is based on a travel diary survey that was planned and executed in cooperation with Vilnius Municipality during preparation of sustainable mobility plan. The following work describes the research object, the questionnaire design, sampling strategy and the analysis of results based on characteristics of respondents. An innovative activity sequence-focused travel behavior research approach designed to collect data for a tour-based travel demand model.

Machine learning applications in activity-travel behaviour research: a review

ABSTRACTThis paper reviews the activity-travel behaviour literature that employs Machine Learning (ML) techniques for empirical analysis and modelling. Machine Learning algorithms, which attempt to build intelligence utilizing the availability of large amounts of data, have emerged as powerful tools in the fields of pattern recognition and big data analysis. These techniques have been applied in activity-travel behaviour studies since the early ’90s when Artificial Neural Networks (ANN) were employed to model mode choice decisions. AMOS, an activity-based modelling system developed in the mid-’90s, has ANN at its core to model and predict individual responses to travel demand management measures. In the dawn of 2000, ALBATROSS, a comprehensive activity-based travel demand modelling system, was proposed by Arentze and Timmermans using Decision Trees. Since then researchers have been exploring ML techniques like Support Vector Machines (SVM), Decision Trees (DT), Neural Networks (NN), Bayes Classifiers, and more recently, Ensemble Learners to model and predict activity-travel behaviour. A large number of publications over the years and an upward trend in the number of published articles over time indicate that Machine Learning is a promising tool for activity-travel behaviour analysis and prediction. This article, first of its kind in the literature, reviews these studies and explores the trends in activity-travel behaviour research that apply ML techniques. The review finds that mode choice decisions have received wide attention in the literature on ML applications. It was observed that most of the studies identify the lack of interpretability as a serious shortcoming in ML techniques. However, very few studies have attempted to improve the interpretability of the models. Further, some studies report the importance of feature engineering in ML-based studies, but very few studies adopt feature engineering before model development. Spatiotemporal transferability of models is another issue that has received minimal attention in the literature. In the end, the paper discusses possible directions for future research in the area of activity-travel behaviour modelling using ML techniques.

Benefits of Integrating Microscopic Land Use and Travel Demand Models: Location Choice, Time Use & Stability of Travel Behavior

The land use/transport feedback cycle has been described since decades. The transport system provides travel times under congested conditions, which are used in land use models to assess the desirability of locations. Based on where people live and work, demand for travel is derived which leads to updated congestion, and hence, new travel times. There is a renewed interest in combined land use and travel demand models, as new trends, such as telework or driverless vehicles, are expected to have substantial influence on land use/transport interactions. Modeling the impact of such scenarios simply cannot afford to leave the land use forecast static and unaffected by changes of the transportation system. The research presented in this paper proposes a new way of integrating a land use model with a travel demand model. Traditionally, land use models provide location and employment information of the synthetic population for the travel demand model, and the transport system feeds back accessibilities as one of many location factors to the land use model. For improving the cause-and-effect correlation of travel demand and land use models an integrated approach is worthwhile. The project, however, will be the first to microscopically integrate these models at the agent level. Individual activity schedules will influence individual household relocation. Vice versa, changes of the housing location, location of work and school places and the likelihood to conduct telework will influence activity patterns for a household. Representing both systems in integrated land use/transport models results in more reasonable sensitivities than provided by models that represent only one domain. What has not been accomplished before is a microscopic integration of travel demand and land use models. Even projects that integrated microscopic land use models with activity-based models did not improve upon the integration step, but merely linked two microscopic models in a traditional way through accessibilities, and thereby, missing out on some of the opportunities offered by microsimulation. This project will integrate the two existing models SILO and mobiTopp. Both are agent-based models and work with synthetic populations that represent the agents microscopically. The microscopic integration will allow for a better representation on land use/transport interactions.

Integrating Household Travel Survey and Social Media Data to Improve the Quality of OD Matrix: A Comparative Case Study

Collecting effective data is a fundamental step in developing transport networks and related research. Social media have become an emerging source of data for traffic analyses. In this paper, we demonstrate that the function of a city influences the utility of social media data in travel demand models by generating models for eight US cities with different functions. Data from Twitter and Foursquare, as well as other socio-demographic information, are considered as independent variables in Origin-Destination trip regression models generated via a Random Forest regression technique. Model performance with and without use of social media data are compared via 10-fold cross-validation. The results indicate that the accuracy of the models for all eight cities improved when independent variables based on social media data were included. The performance was most improved in metropolitan areas, followed by rural and tourist areas. Inspired by this finding, we conclude that the city function influences the utility of social media data in travel demand models. Meanwhile, we create models based on trip purpose and transport mode to explore other factors that may impact the efficiency of applying social media data in transport research.

Mode Choice Analysis using Web-based Revealed Preference Questionnaire, Stated Preference Experiment and Google Maps API

Use of internet for collecting the data is becoming increasingly popular amongst researchers due to its higher speed of data collection and lesser cost since it requires very less workforce. In this study, a webpage based revealed preference (RP) and stated preference (SP) survey is conducted, where the data collected from the revealed preference questionnaire is used in real-time to generate the options for stated preference experiment. For obtaining, the accurate coordinates of the origin-destination of a trip and the travel time by all available modes Google Maps API are used. A mode choice model using RP and SP data individually and a joint RP+SP model using traditional approach is developed and a comparison of the obtained results is made. The developed models suggest that this methodology can be used for collecting the RP data. However, the methodology can be improved for SP experiment and can be helpful in collecting the good quality SP data. Use of Google Maps API provides more accurate travel time and travel distance skims compared to zonal level skims obtained from travel demand model.

Impact of Urban Sprawl on Travel Demand for Public Transport, Private Transport and Walking

By growing of population, the concept of urban sprawl is increasingly recognized as a serious dilemma among the cities, especially in developing countries. Kabul is one of those cities which are sprawling for two decades and the implication is that residents are suffering from long travel time, pollution and energy consumption. In Kabul city, the number of users for public transport and walk are decreasing day by day and they are shifting to other alternatives. A detailed study to be made in understanding urban sprawl and urban compactness influences on choice mode. This paper is aimed to evaluate the impact of urban sprawl and urban compactness on travel demand for transport modes in 22 districts in Kabul city. Travel demand modeling by using the traditional four-step transportation forecasting model for the year 2017 has been applied to assess the current number of passengers for each transport mode (public transport, walk and private transport) in all 22 districts. Secondly, urban compactness based on metrics with two indicators, mixed-use and density measured to figure out the level of sprawl and compactness in each district. Finally, travel demand for transport modes in both circumstances (compact and sprawl areas) compared in observed districts. The extracted outcome presented an interesting vision of variation amongst districts in Kabul city. This study revealed a profound relationship between urban sprawl and travel demand for all urban transport modes.

Exploring the Transferability of FEATHERS – An Activity Based Travel Demand Model – For Ho Chi Minh City, Vietnam

This paper presents descriptive analyses to investigate the transferability of FEATHERS, an activity-based travel demand model developed for Flanders (Belgium), to a region in a developed country to Ho Chi Minh City (Vietnam), a metropolitan area of a developing country, using personal and household travel survey data. The transferability was tested by using indicators which represent individuals’ activity and travel behaviors. The results confirm the transferability of the FEATHERS modelling structure to Ho Chi Minh City (HCMC), yet all sub-models require recalibration due to the different cultural settings between the two study areas. The differences in transport mode options, land use patterns and individuals’ location choice preferences make transport mode and location choice the least transferable models.

Modeling the Socio-Economic Impacts of Covid-19 Using an Improved Traditional Travel Demand Model

The emergence of the COVID-19 virus and subsequent pandemic has caused unprecedented disruption across many aspects of life across the globe. The implications have necessitated a re-evaluation of travel demand models to include the effects that mass isolation and quarantines have on roadway networks. The present paper is focused on data and methodologies to change travel demand models to account for differences encountered during the COVID-19 pandemic. Increased telecommunicating, coupled with store closures and restrictions, within the majority of business categories, has drastically changed traffic patterns. The results indicate that travel has decreased significantly, as businesses have altered their operating practices, requiring employees to work from home due to the restrictions implemented by the government. The present methodology considers a variety of potential alterations to travel patterns resulting from the presence of COVID-19. Because different amounts of reduced employment and increased telecommuting are considered, the present methodology and associated analysis show that different strategies proposed or implemented by the government can be modeled. The results provided in this paper, and the associated model, thus provide a means to account for the new traffic modelling reality resulting from the presence of COVID-19. Such travel modeling is important in regard to forecasts of future travel demand, associated transportation planning, allocation of present and future transportation resources, resulting economic benefits and challenges, and the well-being of our society, as an operationally efficient transportation network, and the associated quality of life, are maintained.

A microeconomic framework for integrated agent-based modelling of activity-travel patterns and energy consumption

Abstract The sophistication in the demand management approaches in both transport and energy sectors and their interaction call for modelling approaches that consider both sectors jointly. For agent-based microsimulation models of travel demand and energy consumption, this implies the necessity to ensure consistent representation of user behaviour with respect to mobility and energy consumption behaviours across the model components. Therefore this paper proposes a microeconomic framework, termed the HOT model (Home, Out-of-home, Travel) grounded in the goods-leisure paradigm, but extended to incorporate emerging activity-travel behaviour patterns and their energy consumption implications. We discuss how the model can be operationalised and embedded within agent-based frameworks with a case study using time use and energy consumption data from the UK.

SIMBA MOBI: Microscopic Mobility Simulation for Corporate Planning

This paper presents a microscopic travel demand model, which simulates the entire population of Switzerland as individual agents. For each agent, all mobility decisions are computed microscopically: from activity generation, tour construction, choice of destinations and travel modes, through route choice and network simulation. The model is called SIMBA MOBi. It has been developed over a period of two years by SBB (Swiss Federal Railways). It simulates all modes of transport over 24 hours of the average weekday. It covers the entire country of Switzerland, including urban, rural and intercity travel, and is applied to support business decisions within the SBB corporation. SIMBA MOBi is one of the first microscopic models that was developed inside a transport operating company in Europe and is applied to support real world decisions regarding service and infrastructure.The paper gives an overview of the model’s architecture, methodology, input data and software. Then, calibration approach and validation results are presented, and an insight into first applications of the model is given. The paper concludes with a discussion of cost and benefits of microscopic modeling in real-world transportation planning.

Assessment of urban freight travel characteristics - A case study of Delhi

India spends 15-20% of its GDP on transport and logistics and Indian freight transport market is expected to grow at a CAGR of about 13% by 2020. Road freight constitutes around 63% of total freight movement and the average speed of trucks on Indian roads is about 20 km/hr covers only 250-300 km a day compared to 700-800 km in developed countries. Moreover, the total trip expenses increases about 15% due to the delays at check-posts by Transport Department and Traffic Police, toll plazas etc. The working conditions of truck drivers also deteriorate as they work for long hours, resulting in high stress and fatigue leading to accidents. The need is recognized for collaboration amongst stakeholders to identify optimal freight policies and pursue a rapid deployment of improvements. In this direction, creating better data and models are needed to enable planners to better predict freight movement and design better informed policies. In the present study, the freight transport data has been exhaustively collected to assess the urban freight characteristics of Delhi. For this purpose, the field surveys namely outer cordon traffic surveys at 10 locations and focal point survey at 20 locations within the city have been conducted. The results can make a good database to understand the freight characteristics of a city and it would be highly useful in developing freight travel demand models and also evaluate appropriate freight related transport policies for their effective implementation on the urban road network by the policy makers and concerned authorities.

Multi-Mode Traffic Demand Analysis Based on Multi-Source Transportation Data

Automatic vehicle identification (AVI) data, Integrated Circuit (IC) card data and Global Positioning System (GPS) data offer an emerging and promising source of information for analysis of traffic problems. Research on insights and information from AVI data for transport analysis has made little progress in developing specific applications especially. The emergence of multi-source data provides us with a new perspective for multi-mode transportation. This paper proposes a multi-mode traffic demand forecasting method based on AVI data, metro IC card data, and taxi GPS data. The paper extracts traffic origins and destinations (OD) information of travelers from the multi-source data and uses the extracted data for traffic zone division. Finally, a multi-mode traffic forecasting model is established on this basis. GPS data of taxi trips are selected as the clustering data and k-means algorithm was adopted to divide traffic zones in Shenzhen. Moreover, the research applies the principle of convex hull to outline the boundary of the cell. Additionally, this paper establishes the multi-mode transportation forecasting model by integrating the correlation between various transportation modes into the deep learning model for prediction. The results show that the multi-mode demand forecasting model has higher accuracy and better forecasting results comparing it with the single-mode demand forecasting model which is referring to the conventional four-step procedure. The result demonstrates that effective traffic and travel data can be obtained from multi-source data, providing an opportunity to improve the analysis of complex travel patterns and behaviors for travel demand modeling and transportation planning. Furthermore, the substantive contribution of this research is that it provides strong empirical evidence for the existence of correlation among multi-mode travels and travel demand.

Updating of a Travel Demand Model Based on Fuel Consumption Data

Traditional methods like Home Interview Survey (HIS) used for estimation of travel demand in an urban area are not only costly and time-consuming, but also suffer from problems like sample bias, significant no-response rate, and data coding errors. In addition, such data collection processes still need to be updated on a continuous basis. Due to the above issues, there is a need for the use of more accurate secondary big data sources for the estimation of travel demand, such as mobile call records, mobile/GPS navigation data, travel smart card, and fuel smart card data.Development of travel demand forecasting models (TDFM)s based on up-to-date variables can help decision-makers achieve efficient solutions to the sudden transport problems, highlighting the need for new tools and methods that are comparatively cheaper, yet more accurate and more efficient. Nevertheless, the validity assessment of combined revealed and stated preference (RP and SP) data for analysis purposes often becomes a complicated task.This study tries to offer an improved and efficient alternative to the conventional and tedious data instruments, like HIS, for the development of a TDFM, where the big data collected through the fuel smart cards (7.5 million records of refueling data), as well as water and electricity consumption data, are used for examining travel demand patterns in the city of Shiraz in Iran. Coupled with other information technologies, smart fuel card data can provide an alternative replacement to the conventional time-consuming transport planning surveys. The large dataset of all the 64 fuel stations, amounting to 7.5 million refueling records for 61 days, was analyzed in this study, and valuable information was obtained that correlated well with traffic parameters for the city.

Exploring strengths and weaknesses of mobility inference from mobile phone data vs. travel surveys

Origin–destination (OD) matrices serve as a basis for travel demand modelling. Traditionally, they are derived from travel surveys that collect detailed trip information but with several shortcomings. Mobile phones are regarded as a useful source of information on people’s daily mobility. This work explores the use of mobile data in the context of mobility studies by comparing matrices derived from both types of sources over the same region. The results reveal many common features in the trip information. Moreover, although the use of mobile technology may raise questions for short trips, the huge representativeness of this technology captures the mobility in OD connections extensively regardless the area. This is crucial for non-populated areas (e.g. industrial parks or educational campuses), which constitute important mobility hotspots. Based on these findings, an applicable data fusion approach to obtain the optimum accuracy from these heterogeneous sources is presented and applied.

Towards agent-based travel demand simulation across all mobility choices – the role of balancing preferences and constraints

This article presents an agent-based travel demand model, where agents react to transport supply across all mobility choices. Long-term choices include mobility tool ownership and work locations. Daily travel patterns are simulated at the individual level by sequentially combining activity frequency, activity durations and destinations as well as a rule-based time-of-day scheduling. A key to success in this novel approach is balancing individual preferences of travelers with system constraints. The model incorporates two types of constraints: 1) capacity constraints of the transport infrastructure. 2) natural time and space constraints during the execution of individual 24-hour day plans. Model results are validated against empirical observations of travel demand in Switzerland. The article concludes with a perspective for further research and development in the field of applied agent-based modeling.

Previsão da demanda por viagens domiciliares através de método sequencial baseado em população sintética e redes neurais artificiais

A estimativa de viagens por domicílio é fundamental para a tomada de decisões relativas ao planejamento de transportes. Porém, para obter essa estimativa são necessários dados desagregados dos domicílios, que geralmente são obtidos pela Pesquisa Domiciliar de Origem e Destino. No entanto, a maioria das cidades enfrenta problemas para a aquisição desses dados, uma vez que este tipo de pesquisa é de alto custo de preparação e execução. Desta forma, surge a necessidade de ferramentas que forneçam dados confiáveis e com baixo custo. Assim, o objetivo deste artigo é apresentar um método sequencial, para estimativa de viagens domiciliares, a partir de população sintética e Redes Neurais Artificiais (RNAs). A população sintética foi baseada em dados agregados do censo e simulação Monte Carlo. Os resultados obtidos com as RNAs foram comparados aos resultados de um modelo linear tradicional, mostrando-se melhores e corroborando o potencial do uso de RNAs para modelagem da demanda por transportes. As viagens sintéticas por domicílio foram validadas a partir dos dados desagregados da Pesquisa Origem-Destino (2007) e testes de hipótese para comparação de valores típicos e distribuições populacionais. Em 71% dos setores censitários, as viagens sintéticas foram consideradas similares aos dados reais, confirmando a eficiência do método proposto. Assim, a principal lacuna desta pesquisa, é a apresentação do método sequencial, capaz de tanto minimizar problemas de aquisição de dados quanto atenuar as restrições e suposições matemáticas, inerentes aos modelos tradicionais de demanda por transportes.

Smart charging management for shared autonomous electric vehicle fleets: A Puget Sound case study

Increasingly, experts are forecasting the future of transportation to be shared, autonomous and electric. As shared autonomous electric vehicle (SAEV) fleets roll out to the market, the electricity consumed by the fleet will have significant impacts on energy demand and, in turn, drive variation in energy cost and reliability, especially if the charging is unmanaged. This research proposes a smart charging (SC) framework to identify benefits of active SAEV charging management that strategically shifts electricity demand away from high-priced peak hours or towards renewable generation periods. Time of use (TOU), real time pricing (RTP), and solar generation electricity scenarios are tested using an agent-based simulation to study (1) the impact of battery capacity and charging infrastructure type on SAEV fleet performance and operational costs under SC management; (2) the cost reduction potential of SC considering energy price fluctuation, uncertainty, and seasonal variation; (3) the charging infrastructure requirements; and (4) the system efficiency of powering SAEVs with solar generation. A case study from the Puget Sound region demonstrates the proposed SC algorithm using trip patterns from the regional travel demand model and local energy prices. Results suggest that in the absence of electricity price signals, SAEV charging demand is likely to peak the evening, when regional electricity use patterns already indicate high demand. Under SC management, EVs with larger battery sizes are more responsive to low-electricity cost charging opportunities, and have greater potential to reduce total energy related costs (electricity plus charging infrastructure) for a SAEV fleet, especially under RTP structure.

A framework to guide strategic disinvestment in roadway infrastructure considering social vulnerability

Abstract Strategic disinvestment involves a conscious reallocation of funds to prioritize transportation infrastructure assets in terms of their importance or criticality, and then deliberately shifting limited resources away from lower priority assets toward higher priority assets. While previous investment paradigms considered the economic implications of expanding or maintaining assets versus the alternative of no expansion or no maintenance, disinvestment considers whether it is realistic to attempt to maintain all assets according to their original standards, or if performance standards should be changed or ownership transferred. While the concept of disinvestment has received attention in recent years, there is no established framework for disinvestment actions, potential travel impacts associated with disinvestment actions have not been widely examined, and it is unclear how disinvestment may affect socially vulnerable populations. In this paper, we present a novel framework to guide disinvestment decisions. The framework identifies different disinvestment actions and illustrates the savings, costs, and control tradeoffs associated with those actions. Each action offers different cost-savings and is associated with varying levels of reduced responsibility or obligation. We next introduce a measure of social vulnerability referred to as the Disinvestment Vulnerability Index (DVI) to help identify socially vulnerable populations impacted by disinvestment. We then propose a methodology for identifying a set of least critical or “non-essential” roadway assets as initial candidates for disinvestment, and introduce an approach for quantifying the potential effect of asset-specific disinvestment actions on vulnerable populations. The suggested approach allows decision-makers to “take the first step” in selecting an initial set of potential disinvestment candidates by identifying assets that have a minimal disruptive effect on network-wide travel, while simultaneously considering accessibility and mobility issues related to socially vulnerable populations. We illustrate the approach in the state of Vermont using the state’s travel demand model and data obtained from the U.S. Census.

Estimating exposure to fine particulate matter emissions from vehicle traffic: Exposure misclassification and daily activity patterns in a large, sprawling region

Vehicle traffic is responsible for a significant portion of toxic air pollution in urban areas that has been linked to a wide range of adverse health outcomes. Most vehicle air quality analyses used for transportation planning and health effect studies estimate exposure from the measured or modeled concentration of an air pollutant at a person's home. This study evaluates exposure to fine particulate matter from vehicle traffic and the magnitude and cause of exposure misclassification that result from not accounting for population mobility during the day in a large, sprawling region. We develop a dynamic exposure model by integrating activity-based travel demand, vehicle emission, and air dispersion models to evaluate the magnitude, components and spatial patterns of vehicle exposure misclassification in the Atlanta, Georgia metropolitan area. Overall, we find that population exposure estimates increase by 51% when population mobility is accounted for. Errors are much larger in suburban and rural areas where exposure is underestimated while exposure may be overestimated near high volume roadways and in the urban core. Exposure while at work and traveling account for much of the error. We find much larger errors than prior studies, all of which have focused on more compact urban regions. Since many people spend a large part of their day away from their homes and vehicle emissions are known to create "hotspots" along roadways, home-based exposure is unlikely to be a robust estimator of a person's actual exposure. Accounting for population mobility in vehicle emission exposure studies may reveal more effective mitigation strategies, important differences in exposure between population groups with different travel patterns, and reduce exposure misclassification in health studies.