Activity-based Travel Demand Model Research Articles

Developing an Integrated Activity-Based Travel Demand Model for Analyzing the Impact of Electric Vehicles on Traffic Networks and Vehicular Emissions

Developing an Integrated Activity-Based Travel Demand Model for Analyzing the Impact of Electric Vehicles on Traffic Networks and Vehicular Emissions

An OpenStreetMap-based approach for generating capacity-restricted POIs for activity-based travel demand modeling

The generation of Points of Interest including capacities for travel demand models is usually costly and laborious. Recently, several methods have been developed to automate this process based on Volunteered Geographic Information data as well as to validate the corresponding results. The methodology presented in this paper takes on the task of setting up such a workflow on a large scale, including capacities and all common activity types, by making use of OpenStreetMap data. As a result, a Points of Interest dataset could be created for the whole of Germany, which maps around 24 million destinations including capacities divided into seven activity types. An exemplary validation based on the generated workplace and shopping locations as well as capacities indicates a varying quality of results depending on the respective regional type. The discussion additionally concludes that assignment of several activities to one location presents challenges that should be addressed in future approaches.

Using App Usage Data from Mobile Devices to Improve Activity-based Travel Demand Models

Using App Usage Data from Mobile Devices to Improve Activity-based Travel Demand Models

Microsimulation of activity generation, activity scheduling and shared travel choices within an activity-based travel demand modelling system

This study presents the microsimulation of activity generation, activity scheduling and shared travel choice processes within an activity-based shorter-term decisions simulator (SDS). Activity generation simulates individuals’ daily activities in an orderly fashion utilizing a Markov Chain Monte Carlo modelling technique within the SDS model. Activity scheduling is simulated as a process of activity agenda formation, destination location choice and shared travel choice decisions utilizing heuristics and econometric modelling (e.g. mixed logit) approaches. The proposed SDS model addresses multi-way interactions among different activity and travel decision components. It also accounts for individuals’ social interactions within the modelling framework. The model is implemented within C#.NET platform for Halifax, Canada from 2006 to 2036 on each year interval. The performance of the model is measured based on absolute percentage error (APE) values for the year 2011, and microsimulation results are validated with the 2016 Canadian Census information. The simulation results suggest that higher density of activity destinations is predicted within approximately 10 km distances from home locations in 2006. Interestingly, with time, the densities become more skewed to the left of 10 km. The highest percentage of non-shared travel (32%) is predicted for a travel length of less than 5 km in 2036, which decreases with distance. In contrast, the proportions of shared travel choice alternatives are predicted to increase as travel length increases. This paper offers critical insights on spatio-temporal evolution of activity and travel decisions, which will be useful for integrated transportation and land use policy making.

On the use of common random numbers in activity-based travel demand modeling for scenario comparison

ABSTRACT Activity-based travel demand models provide a high level of detail when modeling complex travel behavior. Since stochastic simulation is used, however, this high level may induce large random fluctuations in the output, necessitating many model reruns to produce reliable output. This may become prohibitive in terms of computation time when comparing travel behavior between multiple scenarios, in which case each scenario requires its own simulation. To alleviate this issue, we study the use of common random numbers, which is a technique that reuses the same random numbers for choices made by travelers between scenarios. This ensures that any observed difference in output across scenarios cannot be attributed to mutual differences in drawn random numbers, eliminating an important source of random fluctuation. We demonstrate by a numerical study that common random numbers can greatly reduce the number of runs needed, and thus also the required computation time, to obtain reliable output.

Analyzing Potential Impacts of Motorcycle Travel Demand Management Using an Activity-Based Travel Demand Model for Ho Chi Minh City, Vietnam

This paper analyses the potential impacts of motorcycle travel demand management strategies for Ho Chi Minh city, a motorcycle dominant city of Vietnam. Four scenarios on managing the use of motorcycle are analyzed using an activity-based travel demand model called FEATHERS-HCMC. The model is adapted from the FEATHERS model developed for Flanders, Belgium and based on the household travel survey in Ho Chi Minh City. The modeling framework of FEATHERS-HCMC is a combination of the heuristic rules and the econometric models. The simulation results suggest that a strong enforcement on motorcycle use would significantly reduce motorcycle trips and induce the rearrangement of location and time-of-day for daily activities. These results provide empirical evidence that shows several advantages of using activity-based travel demand models comparing to the current four-step models. FEATHERS-HCMC is a promising tool to support decision-making process for assessing sustainable transportation policies, not only for HCMC but for other regions with similar characteristics.

Transportation planning using activity-based travel demand model

According to Bangalore's Transportation Department, the number of transport vehicles in the city will be close to 85.6 lakh two wheelers, 50 lakh automobiles, and 20 lakh transport vehicles by the end of 2023. Two-wheelers account for 58.41% of all vehicles, while Light Motor Vehicles (LMVs) account for 22%, Heavy Motor Vehicle (HMVs) for 11%, and other vehicles account for 8.59%. Therefore, there is a need for efficient transportation planning. From earlier research, it was found that activity-based modeling is more efficient for evaluation. In this study, a method was proposed to develop an activity-based travel demand model for the selected zones of Bangalore city, and work trips were generated from the Global Tech Park. The trips were monitored, and it was found that people came from all around Bengaluru. Four zones in Bengaluru were chosen for this study because they provided the most travel to the Global Tech Park. The four zones were Kengeri, Mylasandra, RR Nagar, and Vijaynagar. The data were collected through individual person surveys, considering the influential parameters in developing person tours. The collected data were analyzed using SPSS software, and models were developed based on parameters such as age, sex, monthly income, distance traveled, active travel, daily travel cost, and vehicle ownership. The results obtained in the form of the models were compared with those of traditional models. In addition, the factors influencing the trips of each individual were studied, and the effects of these factors were analyzed. The results obtained were satisfactory in terms of the R2 value and other testing parameters. From the primary data analysis, it was found that the preference for modal choices of the public increased from public transport (i.e., buses) to private transport (i.e., owned cars) as per the increase in age, income, professions, travel distance, and daily travel cost.

The Impact of Socio-Economic, Land Use, and Travel Related Variables on Escort and Non-Escort Intermediate Stops on Work Tours

Stop frequency prediction model is one of the components of the activity-based travel demand models. Most of the previous studies have considered stops during commutes regardless of their purposes. This approach does not yield the contribution of the explanatory variables to the likelihood of making stops of different purposes. Besides, most of the former studies have been conducted in larger metropolitan areas. This study attempts to cover these gaps by using 2012 travel data of Fargo-Moorhead medium-sized US metropolitan area and classifying stops on work tours into escort, non-escort, and a combination of all stops. The results of logit models indicate that personal characteristics of the commuters do not contribute to the escort stop participation likelihood. In addition, household size variables have a large impact on the likelihood of participating in escort stops and participating in the combined stops on the outbound leg of the commutes. Contrary to several previous studies, the significance and sign of the coefficient of income level vary for different stop purposes. Commuters seemed to be more likely to make more than one non-escort stop close to their workplace on the outbound legs of their commutes. The general results suggest separating the stop purposes yields more illustrative results rather than using one model for the combined stops.

Sustainable mobility strategies and their impact: a case study using a multimodal activity based model

Nowadays, many cities are intending to reduce the use of private vehicles. Governments are incorporating new mobility services and are adapting their parking policies to promote a more sustainable mobility, as both strategies are believed to have the potential to reduce private vehicle use. To understand the effects of these strategies, one needs to be able to model complex travel behaviour up to a very high level of detail. Owing to their flexibility, robustness and ability to model travel activity behaviour on an individual level, activity based travel demand models (ABM) offer a highly suitable methodology for this purpose.In this paper, we employ this methodology to perform a case study in a metropolitan region in the Netherlands which surrounds and includes the cities of Rotterdam and The Hague. This region is of vital economic importance and has a very developed and dense road network. The population of this region is growing, which motivates the ambition to improve its accessibility and move towards sustainable mobility. Therefore, the findings of this study are important to similar regions seeking to do this as well. After setting up a suitable, calibrated ABM able to perform a comprehensive study on the effects of new mobility services and parking policy adaptations in the above-mentioned region, we design seven scenarios to give quantitative answers to policy-related questions on how altering features can reduce the extent to which private vehicles are used for travelling. These features include the availability of mobility hubs (hubs on neighbourhood level where sustainable travel modes are linked), the availability of car/bike sharing services, the availability of ‘Mobility as a Service’ (MaaS) subscriptions, the amount of parking capacity in the region and the parking costs. We also study what the impact would be of an improved public transport service with lowered public transport travel times to and from the city centers, and the impact of an improved cycling network infrastructure with significantly lowered travel times for bike and e-bike travellers.Based on the case study, we find that the introduction of mobility hubs alone has limited impact. However, combining this with making sharing services available to the public through MaaS subscriptions, there is a potential to reduce the number of car trips significantly, while the number of trips undertaken by a more sustainable (shared) e-bike increases as well as the number of so-called multi-modal mode trips (trips undertaken by a combination of various modes). Furthermore, improving the public transport service and micromobility network further increases the potential of mobility hubs in terms of making mobility more sustainable. The case study also shows that limiting parking capacity and increasing parking costs in the city centers is especially helpful for the reduction of vehicle use, leading to an improved car flow.

Evaluating the impact of private automated vehicles on activity-based accessibility in Japanese regional areas: A case study of Gunma Prefecture

Automated Vehicles (AVs) are expected to disrupt the transport sector in the mid- to long-term future, and extensive research efforts have been dedicated to studying its potential implications. However, the existing literature is still limited in the context of Japan. To fill this gap, an integrated travel forecasting approach that combines an activity-based travel demand model and dynamic traffic assignment is estimated and validated to evaluate the impacts of AV-induced travel, and the feedback effects between transport supply and demand. The target area of this study is Gunma prefecture, a regional area in Japan. The planning horizon is set to 2040, where the effects of population decrease are reflected in the scenario settings. Policy scenario analysis that assumes complete substitution of human-driven vehicles with AVs suggests increases between 22.1% and 43.6% in total distances traveled, with a corresponding increase in increased CO2 emissions between 23.6% and 46.0%. While this suggests a worsening network level of service, around 2 to 3 min gains in median activity-based accessibility were also observed, suggesting overall positive effects from AV adoption in the study region.

An Advanced Travel Demand Synthesis Process for Creating a MATSim Activity Model: The Case of Ústí nad Labem

In this study, an activity-based travel demand model of the Ústí nad Labem district (Czech Republic) is created. To do this, an advanced travel demand synthesis process is presented by utilizing the Eqasim framework, which is a pipeline-processing, initial raw data to simulation step. The framework is extensively modified and extended with several algorithms in order to utilize multiple data points for increasing realism in mobility for travel demand models. Two major extensions are provided. First, the pipeline framework is improved to estimate inbound and outbound trips of the study area, comprising a main city and 23 surrounding municipalities. The extended framework assigns synthetic gates for the study area as hubs for the inclusion of inbound and outbound trips. Second, the pipeline framework is advanced to provide a more compatible match of travel destination and activity location state. To do this, the extended framework assigns a capacity for each facility identified for the study area, the expected number of visitors to each facility, and the number of residents in each building. The resulting demand model is presented and the generated trips are evaluated based on locational, transport mode, and sociodemographic characteristics with origin–destination (OD) bundling. Additionally, distribution analyses of the present model are conducted to understand the matching results on a detailed level. The results demonstrate that the present model provides a reasonable output for transport researchers when testing different mobility scenarios and the provided extensions helps them to reduce implausible reflections of the distribution of travel and activity characteristics in household travel surveys while creating demand models, thus increasing realism. Lastly, open-source playground and code repository for further future improvement of synthetic travel synthesis methods are created, which enhances a deep understanding of the preparatory and methodological backgrounds required for complex activity-based simulations in order to inspire transport planners.

Activity-based integrated modelling for assessing COVID-19 impacts on transport operations and emissions

ABSTRACT This study combines an integrated transport, land-use, and energy (iTLE) modeling system with traffic microsimulation model and emission simulator for a holistic analysis of COVID-19 pandemic related changes in traffic flows and emissions. An activity-based travel demand model within iTLE informs pandemic traffic operation scenarios for traffic microsimulation modeling. Link-based simulation outputs inform a finer-grained emission estimation process within a MOtor Vehicle Emission Simulator. Results suggest that the overall network performance improves during lockdown as average delays and queue time decrease by 42.04% and 5.9% respectively compared to pre-COVID condition. Emission results reveal that GHG emissions significantly decrease (64%) in lockdown while it starts increasing gradually in post-pandemic period. Link-based emission analysis indicates that major arterial streets achieve a significant reduction in air pollutant emission. The findings of this study will help transportation planners, engineers, and policymakers to devise effective policies for the improvement of transport operations and emissions.

Modeling and Structuring of Activity Scheduling Choices with Consideration of Intrazonal Tours: A Case Study of Motorcycle-Based Cities

The travel demand prediction of an activity-based travel demand model (ABM) is based on a hierarchical structure of multiple choices related to an individual’s activity scheduling. This structure has, however, not been investigated for motorcycle-based cities. The coarseness of the traffic analysis zoning system combined with mixed land use results in a large proportion of intrazonal trips, which demands model enhancement in ABMs for these cities. Using large-scale household travel survey data from Ho Chi Minh City, a major motorcycle-based city in Vietnam, this study investigated the hierarchical structure for non-work activity scheduling, with consideration of three dimensions: (1) activity starting time, (2) travel mode, and (3) destination choices at the tour level with attention given to the impacts of intrazonal tours. Multinomial logit and nested logit models were adopted for model development. Results showed that work durations in the schedule strongly affected the scheduling of non-work activities. The estimated logsum parameters showed empirical evidence that hierarchy could be different for different activity types. Our findings also suggested a significant impact of intrazonal tours on the structuring and modeling of activity scheduling choices. The validation result indicated that our proposed models’ predictive capability is acceptable.

The implications of Mobility as a Service for urban emissions

Mobility as a Service (MaaS), a user-centric framework for delivering a portfolio of multi-modal mobility services, promises to overcome negative externalities associated with the mobility sector by providing convenience to being multimodal and getting citizens away from using their private vehicles. The current work aims at providing pieces of evidence on the extent to which such a promise can be delivered. To this end, first, the adoption of MaaS and the use of various modes within MaaS bundles are examined using empirical data collected from portfolio and stated adaptation choice experiments. Next, an activity-based travel demand model (Albatross) is employed to simulate activity-travel patterns of travelers in the city of Amsterdam, the Netherlands. The results are linked to an emission model and the impacts on emission levels are compared for various scenarios with different MaaS bundles. It is found that the conservative, balanced, and optimistic scenarios decrease emission levels by 3–4%, 14–19%, and 43–54%, respectively.

A scenario-based stochastic programming approach for the public charging station location problem

This paper presents an integrated framework for the optimal planning of public charging stations for plug-in electric vehicles (PEVs) in urban areas. The framework consists of two main components: (i) an out-of-home charging demand model based on an activity-based travel demand model, and (ii) a public charging station location-allocation model using a scenario-based stochastic programming (SP) approach. In order to capture the dynamic charging behaviour of PEV users, a chi-squared automatic interaction detector (CHAID)-based mixed effects decision tree is induced from multi-day activity diaries. Moreover, because the stochastic error of the micro-simulation approach brings about uncertainty, we adopted a two-stage stochastic mixed-integer programming (TSMIP) model, which measures uncertainty by means of a finite set of scenarios obtained from the derived decision rules underlying PEV charging. The proposed approach is demonstrated for the City of Eindhoven, The Netherlands, and benefits of the stochastic solution are discussed.

Validation of an activity-based travel demand modeling system

ABSTRACT This paper presents the validation analysis of an activity-based travel demand modeling system – shorter-term decisions simulator (SDS). It compares microsimulation results of SDS with observed census and travel-activity survey data for the base year 2006, and forecasting years – 2016 and 2018. Base year validation suggests SDS’s reasonable capability to replicate observed data as indicated by lower absolute percentage error (APE) measures. SDS exhibits satisfactory performance for the forecasting years as well. For instance, Kolmogorov–Smirnov (KS) tests during the activity start time and duration analysis for year 2018 yield smaller D-values than critical values, which indicates a better fit model. Prediction of tour-level mode choice, shared travel choice, and vehicle allocation also exhibit good accuracy (± 4%). Validation results presented in this study demonstrate SDS’s promising forecasting capability so that the model can be utilized efficiently to develop and test alternative land use and travel demand management strategies.

Longitudinal investigation of skeletal activity episode timing decisions – A copula approach

This study proposes a joint modelling approach to simulate episode timing decisions of skeletal activities -- activities that typically take precedence in the scheduling process -- using copulas. The analyses focus on examining the correlation between activity start time and duration while exploring the timing behaviour over a 20-year period between 1996 and 2016 in the Greater Toronto and Hamilton Area. For three types of agent classes (“Worker”, “Student”, “Both”), timing decisions associated with three skeletal activities are investigated: “Work”, “Work Business” and “School”. To obtain mutually exclusive homogeneous subsamples, further segmentation of agents is undertaken considering employment and/or student status, and occupation of the individuals. Parametric models are fitted independently to marginal start time and duration distributions using finite mixture models before estimating copula models for their joint distributions. In the “Worker” class, considerable stability is observed in “Work” activity timing models in all occupation sectors for full-time out-of-home employees, except for the “retail sales/service” sector. “Student” class agents’ “School” timing behaviour is also found to be stable in general for full-time students. More noise is observed in the timing behaviour of “Both” class agents. Given the correlation levels observed, it is concluded that start time and duration variables should be modelled jointly, not independently, within an agent-based, activity-based travel demand modelling framework. This paper demonstrates that it is possible to capture complex behaviour in timing decisions through parsimonious models that only utilize a small number of parameters and recommends exploiting behavioural information accrued over time while developing models.

Beyond air pollution at home: Assessment of personal exposure to PM2.5 using activity-based travel demand model and low-cost air sensor network data

Assessing personal exposure to air pollution is challenging due to the limited availability of human movement data and the complexity of modeling air pollution at high spatiotemporal resolution. Most health studies rely on residential estimates of outdoor air pollution instead which introduces exposure measurement error. Personal exposure for 100,784 individuals in Los Angeles County was estimated by integrating human movement data simulated from the Southern California Association of Governments (SCAG) activity-based travel demand model with hourly PM2.5 predictions from my 500 m gridded model incorporating low-cost sensor monitoring data. Individual exposures were assigned considering PM2.5 levels at homes, workplaces, and other activity locations. These dynamic exposures were compared to the residence-based exposures, which do not consider human movement, to examine the degree of exposure estimation bias. The results suggest that exposures were underestimated by 13% (range 5–22%) on average when human movement was not considered, and much of the error was eliminated by accounting for work location. Exposure estimation bias increased for people who exhibited higher mobility levels, especially for workers with long commute distances. Overall, the personal exposures of workers were underestimated by 22% (5–61%) relative to their residence-based exposures. For workers who commute >20 miles, their exposure levels can be at most underestimated by 61%. Omitting mobility resulted in underestimating exposures for people who reside in areas with cleaner air but work in more polluted areas. Similarly, exposures were overestimated for people living in areas with poorer air quality and working in cleaner areas. These could lead to differential estimation biases across racial, ethnic and socioeconomic lines that typically correlate with where people live and work and lead to important exposure and health disparities. This study demonstrates that ignoring human movement and spatiotemporal variability of air pollution could lead to differential exposure misclassification potentially biasing health risk assessments. These improved dynamic approaches can help planners and policymakers identify disadvantaged populations for which exposures are typically misrepresented and might lead to targeted policy and planning implications.

Modelling of Activity-Travel Pattern with Support Vector Machine

Activity based travel demand modelling involves lot of uncertainty due to the complex and varying decision making behaviour of each individual. This study contributes to the literature by assessing the suitability of Support Vector Machine (SVM) in modelling the activity pattern and travel behaviour of workers. Activity and travel behaviour of workers consists of decision outcomes, which can be modelled as classification and regression problems. SVM is a good classifier and regressor with good testing and learning capability, hence the present study used SVM for modelling. It was found that support vector machine models are well performing to predict the activity pattern and travel behaviour of workers. The SVM models developed in the study predicts the temporal variation of mode wise work activity generation. Prediction of temporal mode share of commuters is advantageous to policy makers to experiment the implementation of temporary Travel Demand Management (TDM) actions effectively.

A hybrid of neuro-fuzzy inference system and hidden Markov Model for activity-based mobility modeling of cellphone users

The aim of this paper is to develop an activity-based travel demand model by receiving cellular network data. Our contribution is to model the uncertainty of human behaviors and also the ambiguity in features affecting users’ activities. We used probabilities to model the first aspect and fuzzy theory to treat with the second; therefore, a hybrid model is proposed based on the Hidden Markov Model (HMM) and Fuzzy Inference System (FIS) such that FIS is used in the emission model of HMM. To show the efficiency of this model, we applied the model to the data collected by Irancell operator and validated the results with four different data sources; labeled data collected from volunteers, ground truth data labeled by an expert, activity-based number of trips generated from/attracted to different regions and reported traffic volume of highways. We have shown that the activity recognition accuracy of the model is 83% and an average error of 5% is obtained when comparing the statistics of the model generated activity plans and the corresponding statistics provided in reports. Generated activity plans are also converted to traffic volumes on transportation network links through MATSIM simulation software and the promising R2 value of 0.83 is observed.