Understanding monthly variability in human activity spaces: A twelve-month study using mobile phone call detail records

https://doi.org/10.1016/j.trc.2013.11.003Get rights and content

Highlights

  • We examine monthly human spatial behaviour over a 12-month period.

  • We use mobile phone call detail records as a novel data source.

  • Our proposed approach is able to capture longitudinal human spatial behaviour.

  • We provide a set of new insights on monthly variances in human spatial behaviour.

Abstract

Human activity-travel behaviour (ATB) is a complex pattern of paths and activities in space and time. Studies indicate that ATB is the construction of daily habitual, weekly, monthly and seasonal routines together with strong variety seeking behaviour. Daily habitual travel patterns are usually taken as a basis, but for transportation planners more knowledge is needed on longitudinal trends in human ATB. Empirical data on prolonged perspective are hard to come by while mobile phone based call detail records could be one means of narrowing this research gap. By implementing this method, the present study attempts to provide new insights on individual monthly spatial travel behaviour. Using call detail records obtained from a set of anonymous mobile phone users, we examined their activity locations and activity spaces for 12 consecutive months. We found modest monthly variation in the number of activity locations, whereas there were great variations in the sizes of individual activity spaces. The monthly variation in individual spatial behaviour is explained up to 17% by seasonality, although the variance is predominantly attributed to individual factors and results indicate significant intrapersonal monthly variability. Findings suggest new avenues for future work on ATB from a longitudinal perspective.

Introduction

The importance of adopting an intrapersonal perspective in the study of human activity-travel behaviour (ATB) has increased in recent years as a result of globalisation, individualisation and the development of mobile technologies (Graham, 1998, Kwan, 2007). These changes have increased the complexity of individual ATB; it is becoming more flexible, fragmented and adventure-seeking (Doherty, 2006, Mokhtarian et al., 2006). Already in developed countries work-related or mandatory travel constitutes approximately one fifth of all trips and up to one quarter of the total distance people travel, while the biggest share of travel is leisure-related (Schlich et al., 2004, McGuckin and Srinivasan, 2005). Frändberg (2008) therefore questions whether the concept of human daily routine in relation to ATB should be revised because long-distance travel for leisure, occupational, or social interaction (Limtanakool et al., 2006) has become a part of the daily activity of many individuals. Hence there is a need to understand more about human ATB from a longitudinal (monthly, annual) perspective.

It is helpful to understand human spatial behaviour and its temporal variability over a prolonged period of time in relation to a myriad of different societal issues. For instance, an ability to assess how intrapersonal and environmental (societal and natural) factors affect the ATB of individuals is essential in that it gives urban planners the tools to deal with urban sprawl and design better neighbourhoods (Calabrese et al., 2013). It also enables transport planners to develop intelligent (public) transportation systems, develop strategies to limit traffic congestion, and to improve accessibility (e.g. Rose, 2006, Tang and Thakuriah, 2012, Miller et al., 2013, Perchoux et al., 2013). However, to date, most studies in this area have been predominantly based on short periods of time relating to data obtained from daily travel diaries. Such an approach could be criticised for neglecting the mid- or long-term perspectives of an individual’s ATB and for not considering that intrapersonal variability may be equal to or even greater than interpersonal variability (Pas and Koppelman, 1987, Schlich et al., 2004, Kitamura et al., 2006). One of the first studies to explore the use of longitudinal data indicates that there are clear temporal patterns in individual ATB (Schönfelder and Axhausen, 2010). Moreover, it seems to be a fundamental element of an individual’s daily life along with intrapersonal variability.

Generally, however, the longitudinal perspective has been neglected due to a lack of data for longer periods. Exceptions exist. For instance, Lanzendorf (2010) analyses key events and the effect of these events on travel behaviour through the use of so-called mobility biographies. But the conventional approach is to collect data using travel diaries or interviews. This approach is expensive, creates a high response burden for the respondent, and the responses can be biased (Golob and Meurs, 1986, Richardson et al., 1995, Axhausen et al., 2007). Mobile phone call detail records (CDR) offer a means of overcoming these drawbacks, and enable us to better understand and map individual mobility patterns and to provide new insights on human spatial behaviour (González et al., 2008, Ahas et al., 2010, Järv et al., 2012). Given the ubiquity of mobile phones in our lives (Townsend, 2000) and the advantages provided by mobile phone-based data (Asakura and Hato, 2004, Bar-Gera, 2007, Herrera et al., 2010), CDR data could be a valuable complement to traditional methods, especially in longitudinal human ATB research.

In light of the above, we herein attempt to extend current understanding of human spatial behaviour from a longitudinal perspective using CDR data. The objective of our study is to reveal the variance in individual monthly spatial behaviour during a longer (one-year) study period and to examine the factors that affect variability. Therefore, two research questions are addressed:

  • (i)

    To what extent does monthly spatial behaviour vary over a period of 12 months?

  • (ii)

    To what extent do interpersonal and intrapersonal factors and seasonality explain monthly variations in human spatial behaviour?

Individual spatial behaviour is measured using computational geometry, and statistical analyses are applied to characterise the variance in human spatial behaviour. Our study is in part exploratory, and unique in the following ways: first, we propose an alternative, yet complementary method that makes use of a mobile phone-based dataset to measure individuals’ ATB; second, a 12-month study period is used; and third, the monthly variability of activity spaces is analysed. The remainder of the paper is organised as follows. In Section 2 we introduce the theoretical and conceptual background to our research approach. We then explain the data, methods, and research design specification in Section 3. We provide our empirical results in Section 4 and discuss our findings, draw conclusions, and suggest directions for future research in Section 5.

Section snippets

Variability in human spatial behaviour

Individual ATB is a complex pattern of paths and activities in space and time and is the outcome of the interconnection between individual factors, interaction with other individuals, and external factors such as the surrounding environment and social structure (Hägerstrand, 1970, Pred, 1984). It is (re)shaped by the socio-economic attributes, as well as the needs, life values, preferences, attitudes, prejudices and habits of individuals (Van Acker et al., 2010). Although predominantly based on

Data

Our CDR data originate from the largest Estonian mobile network operator, EMT, which has over half a million active clients. According to Statistics Estonia (2013), the penetration of mobile phone subscriptions (per 100 inhabitants) in 2009 was 117%, while 95% of Estonia’s 1.38 million inhabitants use mobile phones (EMOR, 2008). The database contains records of all outgoing call activities (CA): calls, short messaging service texts, and Internet and data services initiated by the phone owner.

Interpersonal variability in spatial behaviour

The monthly variation in case of the number of monthly unique activity locations (MAL) appears to have a seasonal pattern, with up to 20% of the variation occurring between the Winter (February) and Summer (July) months (Fig. 4). As expected, the average number of meaningful daily activity locations (DAL) and the concentration of daily life at the 10 most frequented locations (Mtop10) remain steady during the 12 months under study. Monthly variation affects the share of ‘new’ (or yearly visited)

Discussion and conclusions

We studied individuals’ monthly spatial behaviour by adopting a method that can be used in conjunction with traditional data collection methods, namely automatically stored mobile phone-based CDR data. An assessment of the monthly spatial behaviour of individuals was made based on several measures with respect to inter- and intrapersonal factors as well as variation over a 12-month period. Previous studies focused on weekly rhythms of human ATB (Tarigan and Kitamura, 2009) or reveal the impact

Acknowledgements

This research was supported by the EU Regional Development Foundation, Env. Conservation and Technology R&D Program Project TERIKVANT 3.2.0802.11-0043, Estonian Science Foundation Grant IUT2-17, and EITSA. The authors are grateful to the mobile network operator EMT, and Positium LBS in Estonia, for providing the data used in this study. All remaining errors are ours.

References (90)

  • X. Ma et al.

    Mining smart card data for transit riders’ travel patterns

    Transportation Research Part C: Emerging Technologies

    (2013)
  • H.J. Miller et al.

    Developing context-sensitive livability indicators for transportation planning: a measurement framework

    Journal of Transport Geography

    (2013)
  • T. Neutens et al.

    My space or your space? Towards a measure of joint accessibility

    Computers, Environment and Urban Systems

    (2008)
  • C. Perchoux et al.

    Conceptualization and measurement of environmental exposure in epidemiology: accounting for activity space related to daily mobility

    Health and Place

    (2013)
  • M.J. Roorda et al.

    Long- and short-term dynamics in activity scheduling: a structural equations approach

    Transportation Research Part A: Policy and Practice

    (2008)
  • L. Tang et al.

    Ridership effects of real-time bus information system: a case study in the City of Chicago

    Transportation Research Part C: Emerging Technologies

    (2012)
  • K.M. White et al.

    Mobile phone use while driving: an investigation of the beliefs influencing drivers’ hands-free and hand-held mobile phone use

    Transportation Research Part F: Logistics and Transportation Review

    (2010)
  • Y. Yuan et al.

    Correlating mobile phone usage and travel behaviour: a case study of Harbin, China

    Computers, Environment and Urban Systems

    (2012)
  • R. Ahas et al.

    Using mobile positioning data to model locations meaningful to users of mobile phones

    Journal of Urban Technology

    (2010)
  • K.W. Axhausen

    Activity spaces, biographies, social networks and their welfare gains and externalities: some hypotheses and empirical results

    Mobilities

    (2007)
  • K.W. Axhausen et al.

    Fatigue in long-duration travel diaries

    Transportation

    (2007)
  • N.S. Baron

    Introduction to special section: mobile phones in cross-cultural context: Sweden, Estonia, the USA and Japan

    New Media and Society

    (2010)
  • R.N. Buliung et al.

    Exploring spatial variety in patterns of activity-travel behaviour: initial results from the Toronto Travel-Activity Panel Survey (TTAPS)

    Transportation

    (2008)
  • M. Castells et al.

    Mobile Communication and Society: A Global Perspective: A Project of the Annenberg Research Network on International Communication

    (2007)
  • Chetan Sharma Consulting, 2013. US wireless Market Update Q4 2012 and Full Year 2012....
  • M. Dijst

    Two-earner families and their action spaces: a case study of two Dutch communities

    GeoJournal

    (1999)
  • S.T. Doherty

    Should we abandon activity type analysis? Redefining activities by their salient attributes

    Transportation

    (2006)
  • N. Eagle et al.

    Eigenbehaviors: identifying structure in routine

    Behavioural Ecology Sociobiology

    (2009)
  • EMOR, 2008. Bimonthly Telephone Survey: CATI-Bus....
  • European Commission, 2002. Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002...
  • L. Frändberg

    Paths in transnational time-space: representing mobility biographies of young Swedes

    Geografiska Annaler B

    (2008)
  • G.D. Garson

    Univariate GLM, ANOVA, & ANCOVA

    (2012)
  • R.G. Golledge et al.

    Spatial Behavior: A Geographic Perspective

    (1997)
  • T. Golob et al.

    Biases in response over time in a seven-day travel diary

    Transportation

    (1986)
  • M.C. González et al.

    Understanding individual human mobility patterns

    Nature

    (2008)
  • Google, 2011. The Mobile Movement: Understanding Smartphone Users, USA (April 2011)....
  • S. Graham

    The end of geography or the explosion of place? Conceptualizing space, place and information technology

    Progress in Human Geography

    (1998)
  • T. Hägerstrand

    What about people in regional science?

    Papers of the Regional Science Association

    (1970)
  • P. Haggett et al.

    Locational Methods

    (1977)
  • S. Hanson et al.

    Assessing day-to-day variability in complex travel patterns

    Transportation Research Record

    (1982)
  • M. Hirsh et al.

    Dynamic model of weekly activity pattern

    Transportation Science

    (1986)
  • F.E. Horton et al.

    Effects of urban spatial structure on individual behavior

    Economic Geography

    (1971)
  • W. Huang et al.

    Anchor points seeking of large urban crowd based on the mobile billing data

  • B. Iglewicz et al.

    How to Detect and Handle Outliers

    (1993)
  • Isaacman, S., Becker, R., Cáceres, R., Kobourov, S., Martonosi, M., Rowland, J., Varshavsky, A., 2011. Identifying...
  • Cited by (191)

    View all citing articles on Scopus
    View full text