Activity-travel Data Research Articles

Exploring temporal and spatial reliability of individual activity space measurements using massive mobile phone data

AbstractThe concept of activity spaces is widely used in various research fields, but the reliability issues of individual activity space measurements remain underexplored. In this study, we develop a modified individual activity space measurement using massive mobile phone data for a month, called the weighted confidence ellipse (WCE), to quantitatively explore the reliability issues. According to the empirical results, we determine that using individual activity–travel data accumulated over a period of at least 2 weeks to measure individual activity spaces is relatively reliable in time. If considering the independence between workdays and days‐off, individual activity–travel information for day‐offs is necessary to exceed 2 weeks. In addition, the spatial reliability of individual activity space measurements demonstrates obvious spatial continuity and heterogeneity. Spatial location is the key factor affecting individual activity space measurements, and the impacts of the separation effects of residence and workplace may deserve attention, especially in the areas with a single urban spatial structure or urban function. This study can help us to improve knowledge on the temporal and spatial reliability of individual activity space measurements, so as to provide some references for studies on related research issues.

Activity Participation and wellbeing during the covid-19 lockdown in Spain

ABSTRACT On March 14, 2020, the Spanish authorities declared a lockdown to bending the Covid-19 pandemic curve. People were asked to stay at home as much as possible, students were obliged to attend classes on-line, civil servants were also required to tele-work, and any work activity that could be carried out from home was recommended to do so as well. Non-essential activities like theatres, restaurants, gyms, etc., were closed. Only work that could not be carried out at home, grocery shopping, taking care of others, walking the dog, visiting the doctor, do banking, and a few more out-of-home activities were allowed. Out-of-home leisure and social activities were not permitted at all. Such limitations of out-of-home activities could influence the wellbeing of people. The present study aims to analyse the wellbeing of people considering the limitations of out-of-home activities during the lockdown. Data collected through a web-survey from 1,653 participants is studied in this paper. The information analysed is related to the satisfaction of the basic psychological needs (autonomy, competence and relatedness) and positive and negative affect. Activity-travel data and sociodemographic characteristics are also considered. Mann–Whitney U tests, Confirmatory Factor Analyses and Structural Equation models are used to find out differences in the wellbeing of people according to their out-of-home activity characteristics, and how these activity characteristics are associated to the psychological variables studied. Findings suggest that those who mostly spent their out-of-home time working felt slightly better. However, as time working increased, more negative affect and autonomy frustration appeared. On the other hand, more time doing grocery shopping was associated with positive feelings. In general, women felt somewhat worse, and older participants felt a bit better during the lockdown. Employed respondents felt better than their unemployed counterparts. Some lessons are learned to improve urban planning during a pandemic.

A joint model of household time use and task assignment for elderly couples with multiple constraints.

Household time-use patterns are expected to reflect each household member’s daily activity participation and duration with intra-household interactions constrained by multiple budgets. Among various activities, the allocated activity derived from undertaking a household task is studied relatively less in the literature. Who will take an allocated activity is a discrete choice problem of household task assignment, and once a household member is assigned with one household task, other members will have more time to do other activities. To better understand household time-use patterns affected by household task assignment, this paper proposed a joint household-level multiple discrete-continuous extreme value-multinomial logit (MDCEV-MNL) model with multiple constraints. The Karush-Kuhn-Tucker (KKT) method combined with the simulation-based maximum likelihood estimation method is proposed to estimate the proposed model. Based on the household activity-travel data from Beijing of China, the proposed model is customized to explore elderly couples’ time-use patterns with intra-household interactions affected by household task assignment. Following the findings, policy implications are suggested to build an age-friendly society.

An innovative GPS smartphone based strategy for university mobility management: A case study at the University of RomaTre, Italy

Private car dependence has adverse effects on both mental and physical health. Travel Demand Management (TDM) strategies are adopted by public and private organizations to encourage users to switch to sustainable transport for daily commutes. Originally, mobility management only focused on employees, but recently various universities worldwide have started to adopt policies specifically targeted at students. To improve their effectiveness, an innovative approach to TDM based on a GPS smartphone application is tested at University of RomaTre (Italy) on a sample of students. The strategy identifies the integration of all mobility management activities into one smartphone application called IPET (Individual Persuasive Eco-Travel Technology). The IPET enables activity-travel data collection and automatic processing, personalized information provision, use of persuasive technology and implementation of a gamification system. Two questionnaires were conducted before-and-after the measure, to assess its effects on psycho-attitudinal factors. The main objective of the investigation was to analyze individuals’ voluntary behavioral changes with regard to travel mode and everyday behavior following implementation of a personalized travel plan. Data analysis indicated a strong positive change in the level of awareness regarding CO2 emissions and calories burnt after the program implementation. Comparison of GPS data collected in the first and second weeks showed, for those who received a personalized travel plan, an 8.1% shift towards more sustainable transport modes.

The Effects of Weather on University Students’ Activity-Travel Patterns: A Case Study in Batu Pahat, Johore

Weather conditions influencing people’s activity-travel patterns, especially on students that portray more active lifestyles (e.g. conducting outdoor physical activities) than middle-age adult and elderly. However, this effects are still less explored in Malaysia, especially in Parit Raja, Johor. Therefore, this study aims to examine the effects of weather on students’ activity-travel patterns by analysing the association between weather parameters (temperature and humidity) and students’ activity-travel patterns in terms of travel time, number of trips, mode choice, and activity duration with regard to mandatory (e.g. study on campus) and non- mandatory (e.g. leisure and routine) activities. Initially, a total of 110 students were selected randomly among the undergraduate students of Universiti Tun Hussein Onn Malaysia (UTHM). However, only 101 students were participated in this study. A one-day of weekday and one-day of weekend activity-travel data were collected by using stated and revealed preferences approach through travel diary and questionnaire instruments. Temperature and humidity hourly data were obtained from the internet sources. The obtained data was analysed by using SPSS software version 23 that include descriptive statistics and multiple linear regression as the model for analysis method. The results show that the number of trips, activity duration and travel time are much higher in weekday compared to weekend. The students preferred to use cars as their main modes to travel. Temperature significantly affect students’ activity-travel patterns in all activities. Mandatory activity-travel patterns is not affected by the weather attributes but otherwise for non-mandatory activity-travel patterns in terms of number of trips. However, all the effects are considered small, based on the adjusted R 2.

Interrelationships between Mode Choice and Trip-chain Choice decisions: A case of Mumbai Metropolitan Region

Mumbai is one of the major cities in India, where the traffic volume is constantly being increased because of urbanization trend and changes in the socio-economic status of the society. These conditions not only influence commuter’s travel behaviour but also making it too complex. Mode choice and trip-chain choice are the two critical factors influencing this complex travel behaviour. In practice, different activity-based models are using different relationship patterns due to lack of consensus and proper empirical evidence. A few studies focus on the directionality of the trip-chain and mode choice decisions. Hence, this study investigates the hierarchical relationship between these two choice decisions including the influences of socio-demographic characteristics on them. This study uses Structural Equation Models for capturing this multidirectional relationship. This study uses a 15-day activity-travel data and activity-travel information obtained from the survey for defining the choice set. Two separate models are estimated for weekdays and weekends. From the model results, it was observed that the mode choice precedes trip-chain choice during weekdays and mode choice and trip-chain choice decisions are simultaneous during weekends. A number of socioeconomic characteristics also play major roles in influencing the relationships. The model results presented in this study are based on the individual level observations. Hence, for future research, it is necessary to incorporate household interactions and to study the relationship between these choices at the household level.

Beyond Commuting: Ignoring Individuals' Activity-Travel Patterns May Lead to Inaccurate Assessments of Their Exposure to Traffic Congestion.

This research examines whether individual exposures to traffic congestion are significantly different between assessments obtained with and without considering individuals’ activity-travel patterns in addition to commuting trips. We used crowdsourced real-time traffic congestion data and the activity-travel data of 250 individuals in Los Angeles to compare these two assessments of individual exposures to traffic congestion. The results revealed that individual exposures to traffic congestion are significantly underestimated when their activity-travel patterns are ignored, which has been postulated as a manifestation of the uncertain geographic context problem (UGCoP). The results also highlighted that the probability distribution function of exposures is heavily skewed but tends to converge to its average when individuals’ activity-travel patterns are considered when compared to one obtained when those patterns are not considered, which indicates the existence of the neighborhood effect averaging problem (NEAP). Lastly, space-time visualizations of individual exposures illustrated that people’s exposures to traffic congestion vary significantly even if they live at the same residential location due to their idiosyncratic activity-travel patterns. The results corroborate the claims in previous studies that using data aggregated over areas (e.g., census tracts) or focusing only on commuting trips (and thus ignoring individuals’ activity-travel patterns) may lead to erroneous assessments of individual exposures to traffic congestion or other environmental influences.

Multi-day activity-travel pattern sampling based on single-day data

Although it is important to consider multi-day activities in transportation planning, multi-day activity-travel data are expensive to acquire and therefore rarely available. In this study, we propose to generate multi-day activity-travel data through sampling from readily available single-day household travel survey data. A key observation we make is that the distribution of interpersonal variability in single-day travel activity datasets is similar to the distribution of intrapersonal variability in multi-day. Thus, interpersonal variability observed in cross-sectional single-day data of a group of people can be used to generate the day-to-day intrapersonal variability. The proposed sampling method is based on activity-travel pattern type clustering, travel distance and variability distribution to extract such information from single-day data. Validation and stability tests of the proposed sampling methods are presented.

An examination of the accuracy of an activity-based travel simulation against smartcard and navigation device data

Activity-based travel simulators have been experiencing difficulty obtaining high quality activity-travel data and network information, which limits the applicability of the simulator to real world problems. For example, accurate information regarding travel time, link traffic volume and trip distribution is essential for sensitivity analysis using an activity-based travel simulator. Survey data, which relies on respondents’ memories, is typically inaccurate. The recent development of big data engineering has enabled us to use passively collected big data such as from smartcards and navigation devices; their travel time and spatial information is highly accurate. Activity-based travel simulation based on the household travel survey (HTS) can therefore identify inaccuracies in simulated travels by comparing smartcard and navigation device data. This paper aims to examine the accuracy of journeys simulated by an activity-based travel simulator, FEATHERS Seoul (FS), against smartcard and car navigation device data collected in Seoul. The analysis found that the activity-based simulator performs well and reproduces individual travel decisions, as reflected by the overall trip frequency and distance, but it partly fails to correctly reproduce geographical distributions in flexible, non-work trip destinations. The results imply that an activity-based travel simulator needs to improve its incorporation of geographical characteristics using big data engineering to enhance the simulated travel accuracy.

Activity Duration Analysis: Discrete-Time Approach for Recurrent Events, Competing Risks, and Multiple States

Most existing activity-based travel demand models are typically implemented on a tour-based microsimulation framework. Two approaches are available to describe people’s daily decision-making processes on activity and travel in the activity-based model framework: rubber banding and growing scheduling. The growing-scheduling approach requires a series of linked dynamic discrete choices of activity episodes, locations, and travel modes to build incrementally an entire day’s activity-travel patterns for individuals in households. To incorporate explicitly time dependencies into activity episode choice and duration, a hazard-based duration model in a discrete-time framework is introduced; this framework is essentially the same as a discrete choice model with temporal dummies as the simplest form of dynamic discrete choice models. In the application of the discrete–time duration model to activity duration analysis, complex situations of activity-travel data are considered; these include multiple states of origin activity, competing risks of destination activity, and a multilevel structure for recurrent activity episodes within individuals. In addition, a circular, or periodic, variable is introduced as a combination of sine and cosine to model time-of-day effects.

Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data

ABSTRACTGlobal Positioning System (GPS) technologies have been increasingly considered as an alternative to traditional travel survey methods to collect activity-travel data. Algorithms applied to extract activity-travel patterns vary from informal ad-hoc decision rules to advanced machine learning methods and have different accuracy. This paper systematically compares the relative performance of different algorithms for the detection of transportation modes and activity episodes. In particular, naive Bayesian, Bayesian network, logistic regression, multilayer perceptron, support vector machine, decision table, and C4.5 algorithms are selected and compared for the same data according to their overall error rates and hit ratios. Results show that the Bayesian network has a better performance than the other algorithms in terms of the percentage correctly identified instances and Kappa values for both the training data and test data, in the sense that the Bayesian network is relatively efficient and generalizable in the context of GPS data imputation.

Modelling of Work Activity Participation of Workers in Calicut City

This paper aims to explore the work activity participation decision of workers in Calicut city, Kerala, India. Several work related travel patterns are identified from the activity-travel data of the study population. Modelling is done considering the prominent travel patterns in the study area. Activity participation is modelled using binary choice model system, with the choices being either to participate in work activity or not. Various socio-demographic variables are considered for model development. Results indicated that a government employee is more likely to participate in work activity compared to a self employed person. On comparison with males, females are observed to be less probable for work participation.

Taking the Tyke on a Bike: Mothers' and Childless Women's Space-Time Geographies in Amsterdam Compared

This paper investigates the relationship between cycling and mothers' mobility in Amsterdam. Considering that mothers (still) tend to be responsible for transporting children and doing so on a bicycle may increase the difficulties of travel, the city's push for cycling may not suit mothers' mobility needs. Hence, this research aimed to uncover whether mothers' physical levels and experience of mobility by bicycle are by any means inferior to childless women's in Amsterdam. Activity-travel data, collected among thirty-seven women living and working in Amsterdam, informed the results of this research. Throughout a space-time GIS analysis, mothers barely differed from childless women in their amount of and predilection for cycling: mothers generally did not consider the transport of their children on the bicycle as an exertion or safety issue, but rather as a practical means of transport and a pleasurable moment to bond with children. However, mothers and childless women often differed in their travel schedules and purposes, implying that mothers face specific mobility challenges, which evolve as children grow older.

EVALUATING STAGGERED WORKING HOURS USING A MULTI-AGENT-BASED Q-LEARNING MODEL

Staggered working hours has the potential to alleviate excessive demands on urban transport networks during the morning and afternoon peak hours and influence the travel behavior of individuals by affecting their activity schedules and reducing their commuting times. This study proposes a multi-agent-based Q-learning algorithm for evaluating the influence of staggered work hours by simulating travelers’ time and location choices in their activity patterns. Interactions among multiple travelers were also considered. Various types of agents were identified based on real activity–travel data for a mid-sized city in China. Reward functions based on time and location information were constructed using Origin–Destination (OD) survey data to simulate individuals’ temporal and spatial choices simultaneously. Interactions among individuals were then described by introducing a road impedance function to formulate a dynamic environment in which one traveler’s decisions influence the decisions of other travelers. Lastly, by applying the Q-learning algorithm, individuals’ activity–travel patterns under staggered working hours were simulated. Based on the simulation results, the effects of staggered working hours were evaluated on both a macroscopic level, at which the space–time distribution of the traffic volume in the network was determined, and a microscopic level, at which the timing of individuals’ leisure activities and their daily household commuting costs were determined. Based on the simulation results and experimental tests, an optimal scheme for staggering working hours was developed.

Experiment on Activity-travel Survey System based on Scheduling System

To discuss or evaluate certain policies for a smart city (e.g., urban transportation systems), it is effective to develop an agent-based simulation that can reproduce an individual's travel behavior and social interaction. Here, activity-travel data is needed to develop a behavior model. However, it is difficult to collect such data over a long time period due to a heavy burden on subjects of the survey. This study proposes a web system to collect an individual's schedule data easily from travel information. Our proposed system has two key characteristics: 1) travel information (e.g., which route is best at a particular time) is recommended automatically based on the concept of a prism when the user enters a new schedule, 2) researchers can utilize users' schedule information as activity-travel data without conducting a special survey. We tested the system with students as users, who expressed satisfaction with the system's usability as well as operability.

Effects of Urban Spatial Form on Individuals’ Footprints: Empirical Study Based on Personal GPS Panel Data from Rotterdam and Eindhoven Area

It is crucial for urban planners to understand the effects of physical planning on individuals’ behavior. A well-designed built environment allows individuals to engage in activities at shorter travel distances, using environmentally friendly transport modes. This paper examines how individuals’ CO2 footprints co-vary with urban form indicators. GPS technology was used to collect activity-travel data. The data for the analyses were collected between May 2012 and July 2013 covering more than 200 respondents from the Eindhoven and Rotterdam regions, the Netherlands. The data includes personal and household characteristics, detailed travel information, GPS traces and detailed car information, including brand, type and year of production. A Bayesian network was used to extract activity-travel diaries from the collected GPS traces, while a Web-based prompted recall survey instrument was used to validate the imputed data. To identify their living environment, respondents’ home coordinates were matched with map data from the municipality. A two-stage sample selection model was used to estimate the effects of physical planning on individual transport mode choice and daily CO2 emissions.

Semantic Annotation of Global Positioning System Traces

Because of the rapid development of technology, larger data sets on activity travel behavior have become available. These data sets often lack semantic interpretation. This lack of interpretation implies that annotation of activity type and transportation mode is necessary. This paper aims to infer activity types from Global Positioning System (GPS) traces by developing a decision tree–based model. The model considers only activity start times and activity durations. On the basis of the decision tree classification, a probability distribution and a point prediction model were constructed. The probability matrix described the probability of each activity type for each class (i.e., combination of activity start time and activity duration). In each class, the point prediction model selected the activity type that had the highest probability. Two types of data were collected in 2006 and 2007 in Flanders, Belgium (i.e., activity travel data and GPS data). The optimal classification tree constructed contained 18 leaves. Consequently, 18 if–then rules were derived. An accuracy of 74% was achieved when the tree was trained. The accuracy of the model for the validation set (72.5%) showed that overfitting was minimal. When the model was applied to the test set, the accuracy was almost 76%. The models indicated the importance of time information in the semantic enrichment process. This study contributes to future data collection in that it enables researchers to infer activity types directly from activity start time and duration information obtained from GPS data. Because no location information is needed, this research can be easily and readily applied to millions of individual agents.

A Structural Equation Model to Analyze Interrelationship between Activity Participation and Trip-Chaining of Nuclear Household

Travel is derived from the necessity to engage in spatially-separated activities. Previous researches have demonstrated that socio-demographics have effects on activity-travel behavior, but few of them focus on the interactions between activity and trip-chaining, especially the influence of trip-chaining on activity. Based on the activity-travel survey data of Suzhou, China, in 2009, complex interactions between activity participation and trip-chaining behavior of nuclear household are explored using structural equation model. Model estimation results show that trade-off and complementarity exist among different types of activities and trip chains. Besides, trip-chaining generation is deeply affected by activity participation. Subsistence activity negatively affects trip-chaining characteristics, while maintenance and leisure activities positively affect it. Furthermore, feedbacks from trip-chaining characteristics to activity participation do exist. Travel time and the number of trip-chaining have significant effects on activity duration.

Applying Sequence Alignment Methods to Large Activity–Travel Data Sets

Sequence alignment methods have recently found increasing application in transportation research for segmentation and classification analysis. Because these methods account for the sequential nature of activity–travel diary data, they are valuable for classifying activity–travel diary data. However, computation time requirements are overwhelming, implying that large-scale applications of these methods have not yet been possible. The performance and applicability of a heuristic approach was explored by using smart card data from Seoul, South Korea. Results were promising and suggest that the proposed heuristic approach could be used to perform segmentation analysis of large activity–travel data sets. Avenues for further improvement of the suggested method are discussed.

Semiautomatic Imputation of Activity Travel Diaries

The new generation of dynamic activity-based models requires multiday or multiweek activity–travel data. Global Positioning System (GPS) tracers may be a powerful technology to collect such data, but previous applications of this technology to collect data of full activity travel patterns (not just time, route, and location) still required a substantial amount of manual data imputation and processing and hence are still demanding for both respondent and researcher. A semiautomatic data imputation system would be a major breakthrough and would involve less respondent burden. This paper reports and illustrates the design of a system called TraceAnnotator that processes multiday GPS traces semiautomatically. The process of imputing transportation modes, activity episodes, and other facets of activity travel patterns is based on a learning Bayesian belief network (BBN), which represents the multiple relationships between spatial, temporal, and other factors, including errors in the technology itself. Activity type is identified by fusing GPS data with geographic information system land use data and personalized land use data. Land use data are built during the data collection process using reverse geocoding and an Internet-based prompted recall survey, which also allows checking and correction of any imputation errors. The prompted recall data are used to update the conditional probabilities of the BBN. Consequently, that the system can learn over time implies that imputation accuracy will improve over time, reducing respondent and researcher burden. A pilot study is presented and potential improvements of the learning algorithm are discussed.