Trip Purpose Research Articles

Trip Purpose Imputation Using GPS Trajectories with Machine Learning

We studied trip purpose imputation using data mining and machine learning techniques based on a dataset of GPS-based trajectories gathered in Switzerland. With a large number of labeled activities in eight categories, we explored location information using hierarchical clustering and achieved a classification accuracy of 86.7% using a random forest approach as a baseline. The contribution of this study is summarized below. Firstly, using information from GPS trajectories exclusively without personal information shows a negligible decrease in accuracy (0.9%), which indicates the good performance of our data mining steps and the wide applicability of our imputation scheme in case of limited information availability. Secondly, the dependence of model performance on the geographical location, the number of participants, and the duration of the survey is investigated to provide a reference when comparing classification accuracy. Furthermore, we show the ensemble filter to be an excellent tool in this research field not only because of the increased accuracy (93.6%), especially for minority classes, but also the reduced uncertainties in blindly trusting the labeling of activities by participants, which is vulnerable to class noise due to the large survey response burden. Finally, the trip purpose derivation accuracy across participants reaches 74.8%, which is significant and suggests the possibility of effectively applying a model trained on GPS trajectories of a small subset of citizens to a larger GPS trajectory sample.

Investigating what makes people walk or cycle using a socio-ecological approach in seven European cities

Large efforts and investments have been made into public transport, walking, and cycling in cities around Europe. Yet, cars remain the most ubiquitous mode of travel in urban areas. Often, research into the barriers facing active travel evaluates only one part of the problem, such as a person’s surrounding environment (the macro level), socially embedded practices that define the activity (meso level), or a person’s own beliefs and sense of identity (micro level). However, barriers and enablers to active travel exist on multiple levels, and interventions to increase walking and/or cycling are less likely to work when implemented in isolation. Hence, a multilevel socio-ecological model is developed to demonstrate and test the importance of assessing these barriers together, and identify interrelationships among them. Using the Physical Activity Through Sustainable Transportation Approaches (PASTA) dataset on the travel behaviour of people in seven different European cities, this paper identifies the constructs that correlate with active travel most. Within PASTA, psychosocial constructs influence the decision to take a trip by bicycle or walk more than built environment variables. In addition, trip purpose and the meso level influence the importance of built environment and attitudinal variables in explaining active travel. These relationships do not vary significantly between cities. This research further supports the use of multi-faceted interventions to increase walking and cycling, rather than focussing on a single policy.

Comprehensive data validation of a combined weekly time use and travel survey

A well-known problem in Household Travel Surveys (HTS) is item-nonresponse, which occurs when complete trips are not reported or only partial information for trips is given. This article presents a comprehensive data validation effort (data completion and correction through immediate call-backs of the survey participants) on travel and non-travel information as part of a one-week Mobility-Activity-Expenditure Diary Survey. Validation recovered 2.5 percent reporting days of mobile persons with complete information on trips and increased the number of trips of pre-existing mobile reporting days by 3.5 percent. The characteristics of these trips confirm and extend the findings from item-nonresponse studies without validation: The majority of the new trips begin in the afternoon; they are mainly short and irregular; the most important trip purposes are home, work, shopping, and leisure. 37 percent of the new trips generated completely new tours; females seem to under-report less than men do. The analysis of the information on the time use (activities) between trips showed that the data was recorded with high accuracy and completeness even before validation. This study confirms typical item-nonresponse patterns and provides a solution for their mitigation through direct follow-up validation. The validation effort increased the time and cost of fieldwork by 12 percent.

Future scenarios for the European railway sector: A methodology based on past trends and a stated preference survey

The strengthening of the railway mode is one of the main policies that can help to achieve more sustainable future mobility. A methodology is presented to estimate future transport demand scenarios over the medium (2030) and long term (2050) considering three trip lengths: urban, regional, and long distance for two trip purposes: work and leisure. The objective of the scenarios is to answer the question: would railways benefit from a 10% drop in car usage? Qualitative techniques have been applied, based on in-depth interviews and focus groups, as well as a stated preference survey where the people sampled answered to different scenarios which included the presence of a new train, with a high level of service, and other possible new modes of transport such as autonomous cars and buses. The results obtained using Best-Worst discrete choice models show that the rail mode could capture a significant part of the demand from the car over all three journey lengths, but especially for long distance, if the new rail mode were introduced.

Spatiotemporally heterogeneous willingness to ridesplitting and its relationship with the built environment: A case study in Chengdu, China

The capability of ridesplitting service to address current urban transportation problems has attracted considerable research interest in recent years. Given that ridesplitting needs an adequate user base to realize “sharing”, its success highly depends on a comprehensive understanding of people’s willingness of using it. However, previous studies mainly focused on the ridesplitting willingness based on questionnaire surveys at the individual level, which may help transportation network companies find potential users, but lacked studies on people’s ridesplitting willingness from the perspective of the built environment and time-space integration, because the macro urban elements are difficult to be perceived by individuals and thus hard to reflect in questionnaires. Thus, this study estimates the ridesplitting willingness rate of different areas in a city at different times of the day by building a model on the shared order rate and shareability from the real-world DiDi Chuxing dataset, using Chengdu, China, as a case study. The spatial lag model (SLM) is further utilized to exam the relationship between the willingness rate, built environment and transportation-related variables. Results revealed the ridesplitting willingness rate has a significant spatiotemporal pattern between the urban centre and urban periphery and can be divided into morning peak, noon plateau, afternoon valley, night peak, and midnight valley. SLM models indicate that the ridesplitting willingness rate has a significant spatial dependency on its vicinity areas at the origin of the daytime periods or at the destination at night. The distance to urban centre, the distance to railway station, travel demand, accessibility to public transit, land use mixture, and the trip purpose are found to be the most relevant variables with the willingness rate, while the points of interest (POIs) are almost irrelevant. The findings of this study can be helpful to the future promotion of this sustainable service and formulation of policy to improve service quality.

The Impact of Hotel Room Colors on Affective Responses, Attitude, and Booking Intention

ABSTRACT Drawing on color psychology, two scenario-based experiments were conducted to investigate how hotel room colors affect consumer’s affective responses, attitude, and booking intention. The moderating effects of trip purpose and consumer characteristic (environmental consciousness) were also examined. In Study 1, consumers demonstrated more positive attitude and higher booking intention for a hotel room that featured the cool color scheme than a room that featured the warm color scheme. Meanwhile, feeling of pleasure mediated the effect of hotel room color on booking intention. In Study 2, we found that consumers who are more environmental conscious demonstrated more positive attitude and higher booking intention for a hotel room featuring the green color than a room featuring the red color. Implications and future research are discussed.

Analyzing travel behavior in Hanoi using Support Vector Machine

ABSTRACT This study investigates travel decisions (i.e. travel mode and destination) in Hanoi (Vietnam) using Support Vector Machine (SVM). First, a travel interview survey was conducted and 311 responses were collected across Hanoi. Second, a SVM model was trained to predict travel decisions and compared with a multinomial logit (MNL) model (as a benchmark). Third, the most important variables that affect travel decisions were ranked and discussed. The results show that SVM achieves an accuracy of 76.1% (compared to 72.9% for MNL). Moreover, proposed parking charges, household income, trip mode, and trip cost are found to be the most important variables. In contrast, trip purpose, gender, and occupation are found to negatively affect the model. Overall, low travel cost and low motorcycle parking charges, especially for commuters and shoppers, make people less willing to switch to more sustainable modes such as public and active transport.

Catch-up immunization for adolescents and young adults during pre-travel consultation in Japan.

Rubella and measles outbreaks in adults occur because of unimmunized or partially immunized status. Travel clinics play an important role in catch-up measles, rubella, mumps, and varicella immunization for adults. We evaluated the need for catch-up measles, rubella, mumps, and varicella immunization by young adults at our travel clinic. This retrospective observational study was conducted at the National Center for Global Health and Medicine from June 1, 2017 to May 31, 2018. Adults aged 16–49 years who received pre-travel consultation and had childhood immunization records were included. Individuals who fully or partially received planned measles, rubella, mumps, and varicella catch-up immunization were classified as “immunized.” We calculated the proportion of “immunized” individuals and analyzed the factors associated with catch-up measles, rubella, mumps, and varicella immunization at pre-travel consultation using logistic regression analysis. Overall, 3,456 individuals received pre-travel consultations during the study period; 827 (336 men, median age 22 years) had childhood immunization records. The most common trip purposes were study (33%) and tourism (24%). The most common destination was Asia (39%). Catch-up immunization of any measles, rubella, mumps, and varicella vaccine was needed by 755 individuals. After consultation, 20–46% of these participants who needed catchup immunization received at least one dose of immunization. Factors that are negatively associated with measles, rubella, mumps, and varicella catch-up immunization were tourism (odds ratio 0.37 to 0.58), yellow fever vaccination (0.45 to 0.50) (excluding varicella), and each disease history (0.13 to 0.40) (excluding rubella and varicella). Further studies are needed to identify barriers to catch-up immunization.

Travel-based multitasking behaviour in Singapore: Determinants and impacts on money-time-seat trade-offs

Investigating travellers’ travel-based multitasking behaviour broadens the horizons of transport policies regarding improving transport services and promoting more sustainable travel behaviour. This study is conducted in Singapore that examines travel-based multitasking preferences and quantifies the impacts on travellers’ monetary trade-offs between travel time savings and seat availability. Considering activity types and duration, five groups of in-vehicle activity are classified from a list of 26 self-reported activity library through performing an exploratory factor analysis. The choice of multitasking is modelled by considering impacts from the perspectives of travel modes, trip characteristics and socio-demographics. Differences across trip purposes namely home to work (HW), work back to home (WH) and leisure/recreation trips (LR) are captured. Public transport riders are found more likely to participate in e-activities for leisure purpose. Car drivers are found to be less likely to perform any multitask during commute, except for social activity on LR trips. Travellers’ monetary trade-offs vary across in-vehicle activities. Herein, people with passive or information- and communication-technology (ICT) related activities are found to be less sensitive to changes in travel time during WH and LR trips. Carrying out ICT-related activities is found to reduce individuals’ economic cost on morning commute, but to increase it on evening commute. People with productive activity are found slightly more likely to pay for a seat during morning commute, and less likely to do so during evening commute. In effect, we contend travel-based multitasking has a significant, but largely unaccounted for, impact on transport planning and appraisal process.

ActivityNET: Neural networks to predict public transport trip purposes from individual smart card data and POIs

ABSTRACT Predicting trip purpose from comprehensive and continuous smart card data is beneficial for transport and city planners in investigating travel behaviors and urban mobility. Here, we propose a framework, ActivityNET, using Machine Learning (ML) algorithms to predict passengers’ trip purpose from Smart Card (SC) data and Points-of-Interest (POIs) data. The feasibility of the framework is demonstrated in two phases. Phase I focuses on extracting activities from individuals’ daily travel patterns from smart card data and combining them with POIs using the proposed “activity-POIs consolidation algorithm”. Phase II feeds the extracted features into an Artificial Neural Network (ANN) with multiple scenarios and predicts trip purpose under primary activities (home and work) and secondary activities (entertainment, eating, shopping, child drop-offs/pick-ups and part-time work) with high accuracy. As a case study, the proposed ActivityNET framework is applied in Greater London and illustrates a robust competence to predict trip purpose. The promising outcomes demonstrate that the cost-effective framework offers high predictive accuracy and valuable insights into transport planning.

Understanding factors influencing shared e-scooter usage and its impact on auto mode substitution

• A survey is conducted to obtain sociodemographic information of shared e-scooter users, user opinions and riding behaviors. • Random parameter order probit model is estimated to understand use frequency of shared e-scooters. • Mixed logit model is used to model the mode substitution of shared e-scooters. • Shared e-scooters can address congested parking issues and are promising for reducing auto use. • Recommendations on improving shared e-scooter programs are presented. Shared e-scooter is a fairly new transportation mode that emerged in late 2017. Since then, it has gained popularity around the world; however, it also has spiraled into disarray in many cities due to the lack of e–scooter regulations and improper parking and riding behaviors. Limited understanding of shared e–scooters restrains policymakers from developing more effective regulations and promoting this sustainable transportation mode. This study takes a step towards understanding e-scooter user behaviors by investigating factors that influence e-scooter sharing usage and auto mode substitution. Survey data were collected from shared e-scooter users, and random parameter models were applied to explore the factors influencing e-scooter sharing usage and mode substitution. Factors considered in models include sociodemographic information, user behaviors, trip purposes, and health indicators. Model results identify several factors that significantly influence shared e-scooter usage, including user gender, helmet use, exposure to shared e-scooters, ownership of an e-scooter, riding locations, opinions on speed limits, and trip purposes. Contributing factors for auto substitution suggest that shared e-scooters potentially are competing with TNC/taxi, lower cost, and social/entertainment trip purpose, and user households with multiple vehicles contribute to private vehicle substitution. Research outcomes suggest that shared e-scooters could play a significant role in urban transportation sustainability . Insights related to better practices for e-scooter regulations and planning are discussed to help cities improve shared e-scooter program performance and make it a more sustainable transportation mode.

Factors influencing dock-less E-bike-share mode substitution: Evidence from Sacramento, California

Dock-less e-bike-share use is likely to reduce vehicle miles traveled (VMT) and related greenhouse emissions – if it substitutes for car use. If the major mode shift comes from public transit, owned bike, or walking, the benefits will be more limited. The goal of this paper is to identify the factors influencing mode substitution, defined here as the mode that is replaced when bike-share is used. The analysis shows that the dominant mode substitution for trips of less than 1 mile for most trip purposes is walking. Long trips and non-commute trips that start at non-commercial locations are likely to represent car substitution and some groups, such as women, non-membership holders and those who have a private car, are more likely to report car substitution for any trip purpose. These results provide guidance for designing bike-share operations and policies to enhance car substitution.

Big data for big issues: Revealing travel patterns of low-income population based on smart card data mining in a global south unequal city

Smart card data (SCD) allow analyzing mobility at a fine level of detail, despite the remaining challenges such as identifying trip purpose. The use of the SCD may improve the understanding of transit users' travel patterns from precarious settlements areas, where the residents have historically limited access to opportunities and are usually underrepresented in surveys. In this paper, we explore smart card data mining to analyze the temporal and spatial patterns of the urban transit movements from residents of precarious settlements areas in São Paulo, Brazil, and compare the similarities and differences in travel behavior with middle/high-income-class residents. One of our concerns is to identify low-paid employment travel patterns from the low-income-class residents, that are also underrepresented in transportation planning modeling due to the lack of data. We employ the k-means clustering algorithm for the analysis, and the DBSCAN algorithm is used to infer passengers' residence locations. The results reveal that most of the low-income residents of precarious settlements begin their first trip before, between 5 and 7 AM, while the better-off group begins from 7 to 9 AM. At least two clusters formed by commuters from precarious settlement areas suggest an association of these residents with low-paid employment, with their activities placed in medium / high-income residential areas. So, the empirical evidence revealed in this paper highlights smart card data potential to unfold low-paid employment spatial and temporal patterns.

MODELING PUBLIC TRANSPORT USERS’ TRIP PRODUCTION IN HAWASSA CITY, ETHIOPIA

Today, overcrowded public transport demand, resulting in huge costs in an urban area. Similarly, there are a lot of people who use public transport in Hawassa city. This study aimed to develop public transport users' trip production models at the household level. Some socio-economic characteristics and trip detail of the public transport users were collected randomly from the different households through a questionnaire survey. The data gathered was fed into IBM SPSS package version 20 to develop linear regression models. The developed models are associated with trips for purpose and time intervals of trips made. The developed linear regression models, general trips, work trips, educational trips, and trips made before 8:00 AM and after 4:00 PM had good explanatory power. The value of explanatory power comprised of 0.656, 0.722, 0.549, 0.610 and 0.510. These values indicated the explanation power of the socio-economic characteristics on the trips made. It means the daily trips production was significantly affected by the number of working individuals, the different age brackets, cars and motorcycles, and the monthly income per household. The most frequent public transport users’ trips production regarding the trip purpose and time are work trips and occurred after 4:00 PM. This scenario represented a good model developed in this study. Hence, it is suggested that Hawassa city’s traffic management office use the developed models to predict the future trips demand to provide a proper scheme to avoid congestion during the peak hour of the day.

The level and determinants of multimodal travel behavior: Does trip purpose make a difference?

Multimodality refers to the phenomenon of using more than one mode of transport in a given period. Encouraging multimodality potentially provides an effective solution to reduce CO2 emissions and induce modal shifts toward sustainable transport. This research investigates the extent to which the level and correlates of multimodality differ by trip purpose. We used one-week travel diaries of the English National Travel Survey. Our analyses showed that the level of multimodality varied by trip purpose and the associated time-space variability as well as by the number of trip stages. We found that the level of variability in departure time and travel distance was greater for leisure trips than for maintenance trips, which was in turn greater than for work trips. Trips that were more variable in departure time and travel distance showed on average higher levels of individual multimodality, but only if sufficient stages (at least 3) were made. Moreover, we detected cross-purpose disparities in correlates of multimodality in terms of significance and variance explained. This research may provide support to the development of trip purpose-specific policies aiming to increase multimodality.

Walking, cycling, and public transport for commuting and non-commuting travels across 5 European urban regions: Modal choice correlates and motivations

The objective of this study was to explore individual and contextual-level characteristics associated with active (walking and cycling) and public transport as main travel modes for both non-commuting and commuting purposes, in residents of five European urban regions. We also described participant-reported motivations for modal choice for each journey purpose. The study used multilevel models to investigate cross-sectional associations of individual (i.e. age, gender, educational level) and contextual (defined by a combination of residential neighbourhood characteristics in typologies) characteristics with the choice of active and public transport as outcome. Based on an online survey of 6037 residents of Ghent and suburbs (Belgium), Paris and inner suburbs (France), Budapest and suburbs (Hungary), the Randstad (including the cities of Amsterdam, Rotterdam, The Hague and Utrecht in the Netherlands) and Greater London (United Kingdom), we observed associations with both individual and contextual characteristics.Results of the multilevel modelling show that the probability of using active or public transport as main mode varies depending on both individual and contextual characteristics. At individual level, relations with gender, age, education, weight status and having at least one child varied according to main transport mode and/or purpose. For example, overweight participants reported lower level of cycling for commuting and non-commuting travels than normal-weight participants. In the context of non-commuting travels, participants with one or more child reported less public transport use and more walking (vs participants without children). Among contextual-level variables, urban characteristics of the residential neighbourhood defined by four clusters (according to food environment, recreational facilities and active mobility opportunities) were associated with public transport and walking but not with cycling. For active transport the most important reasons were “I like to travel (on foot or by bike)” and “I want to be physically active” for both travel purposes. “Public transport facilities nearby” was indicated as the most important reason for public transport (for both trip purposes) – the second was “Journey time”.Our findings highlight the importance of exploring a combination of multiple correlates at individual and contextual level according to journey purposes and suggest that the role of health-related individual characteristics such as weight status need further exploration.

Transformation of urban mobility during COVID-19 pandemic – Lessons for transportation planning

IntroductionThe coronavirus disease (COVID-19) pandemic is a global threat that started in Wuhan, China, in 2019 and spread rapidly to the globe. To reduce the spread of the COVID-19, different non-pharmacological control measures have been conducted in different countries, which include social distancing, distance working, and stay-at-home mandates. These control measures had affected global transportation and mobility significantly. This study investigated the short-term changes in urban mobility, tropospheric air pollution, and fuel consumption in two major cities of Saudi Arabia, namely, Riyadh and Jeddah. MethodsIn this study, the dynamics of the number of trips and trip purposes in different provinces of the country were analyzed, focusing on the pandemic period and the lockdown program. These changes impacted fuel consumption and, consequently, air pollutants. The quantity of fuel consumption and its trend was projected considering a few possible fuel consumption and emission scenarios. It is also expected that fuel price plays a role in fuel consumption. The spatial and temporal distributions of the remote sensed tropospheric Nitrogen Dioxide (NO2) levels in different provinces were presented to depict the short 19 and long-term impact on the air quality due to the changes in mobility. ResultsThe significant reduction in urban mobility has been observed since the beginning of the first partial curfew in March 2020 compared to that in 2019. The air pollutant levels (such as NO2) in 2020 after the pandemic were generally less than those of 2019. The fuel consumption has been following a decreasing trend in 2020 starting from January due to dynamic fuel price and the additional influence of pandemic. Based on the current online shopping pattern, it is argued that there will be some permanent behavioral changes in urban mobility, which will decrease some shopping trips at least immediately after the recovery from the pandemic. ConclusionsThis study concluded that the availability of global urban mobility data, remote sensed based tropospheric air pollution data, and global fuel consumption database are important sources of information to investigate the impact of COVID pandemic, especially for the developing countries which suffer from scarcity of pertinent urban mobility information. It seems that, at least in the study area, the spread of COVID-19 is a complex phenomenon in which several exogenous factors, in addition to the curfew protocols, affect the spread of the virus.

Modelling heterogeneous preferences for nature‐based tourism trips

AbstractThis paper studies individual preferences for place‐based attributes in the context of nature‐based domestic tourism trips. We examine the regional characteristics that explain tourist destination choice focusing on taste heterogeneity for distance and temperature. We examine the influence of a set of mean shifters in the marginal utilities for regional amenities, and how individuals are willing to trade distance in exchange for warmer (cooler) climates (i.e., marginal rates of substitution). Using a rich dataset of trips within Spain for nature‐based purposes, we find large heterogeneity in preferences for temperature differentials and distance, with trip purposes acting as moderators.

Departure Time Choice in Schedule-Based Transit Assignment

This paper investigates existing departure time models for a schedule-based transit assignment and their parametrization. It analyzes the impact of the temporal resolution of travel demand and suggests functions for evaluating the adaptation time as part of the utility of a path. The adaptation time quantifies the time between the preferred and the scheduled departure times. The findings of the analysis suggested that travel demand should be discretized into intervals of 1 min, with interval borders right between the full minute, that is, ±0.5 min. It was shown that longer time intervals led to arbitrary run volumes, even for origin–destination pairs with just one transit line and a fixed headway. Although a linear relationship between adaptation time and adaptation disutility is a common assumption in several publications, it cannot represent certain types of passenger behavior. For some trip purposes, passengers may be insensitive to small adaptation times, but highly sensitive to large adaptations. This requires a nonlinear evaluation function.

Fully automated vehicles: A cost-based analysis of the share of ownership and mobility services, and its socio-economic determinants

• Cost-based modelling of the share of different modes of vehicle use. • We consider: manual and autonomous vehicle (AV) ownership and on-demand AV services. • Heterogeneity in travel and cost pattern considered by using representative data. • Ownership (manual or automated) is the least cost option for most cases. • High income, business use, rurality and primary vehicle are associated with ownership. A major uncertainty in the diffusion of autonomous vehicles is the split between ownership and automated mobility services. We calculate total cost of ownership and use (TCOU) to compare four alternatives: private manually driven vehicles, private automated vehicles, automated on-demand exclusive use mobility services (ridesourcing) and automated on-demand pooled mobility services (ridesplitting), for both conventional and electric propulsions. We also included potential usefulness of travel time in the driverless vehicles in the TCOU calculations, thus including some of the non-financial factors in travel decision making. While nearly all current studies are limited to comparing average representative trips or mileage of existing vehicles, we calculate TCOUs for every vehicle in the UK National Travel Survey dataset, considering heterogeneity in mileage patterns, trip purposes, time spent driving, value of time, vehicle age, depreciation and other factors between different vehicles. The results suggest that a near-total transition to automated ride services is highly unlikely, since ownership of (automated or manually driven) vehicles continues to be the least-cost option in most cases. Even in the most pro-mobility service test case, ownership remains more cost-effective for one third of the current vehicle fleet. Regression analysis shows that higher income of the main driver, business use of the vehicle, rural location of the household or being the main household vehicle leads to a higher likelihood that automated vehicle ownership will be lower cost compared to automated mobility services. Within automated on-demand ride services, exclusive use services are cost effective for more cases compared to pooled, shared-use type options, with uncertain consequences for future travel demand.