Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
. 2014 Jun 16;9(6):e96180.
doi: 10.1371/journal.pone.0096180. eCollection 2014.

Exploring universal patterns in human home-work commuting from mobile phone data

Kevin S Kung  1 , Kael Greco  2 , Stanislav Sobolevsky  3 , Carlo Ratti  2
Affiliations
Comparative Study

Exploring universal patterns in human home-work commuting from mobile phone data

Kevin S Kung et al. PLoS One. .

Abstract

Home-work commuting has always attracted significant research attention because of its impact on human mobility. One of the key assumptions in this domain of study is the universal uniformity of commute times. However, a true comparison of commute patterns has often been hindered by the intrinsic differences in data collection methods, which make observation from different countries potentially biased and unreliable. In the present work, we approach this problem through the use of mobile phone call detail records (CDRs), which offers a consistent method for investigating mobility patterns in wholly different parts of the world. We apply our analysis to a broad range of datasets, at both the country (Portugal, Ivory Coast, and Saudi Arabia), and city (Boston) scale. Additionally, we compare these results with those obtained from vehicle GPS traces in Milan. While different regions have some unique commute time characteristics, we show that the home-work time distributions and average values within a single region are indeed largely independent of commute distance or country (Portugal, Ivory Coast, and Boston)-despite substantial spatial and infrastructural differences. Furthermore, our comparative analysis demonstrates that such distance-independence holds true only if we consider multimodal commute behaviors-as consistent with previous studies. In car-only (Milan GPS traces) and car-heavy (Saudi Arabia) commute datasets, we see that commute time is indeed influenced by commute distance. Finally, we put forth a testable hypothesis and suggest ways for future work to make more accurate and generalizable statements about human commute behaviors.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist. We acknowledge that Audi Volkswagen, BBVA, The Coca Cola Company, Ericsson, Ferrovial, GE provided funding for this study. There are no patents, products in development or marketed products to declare. This does not alter the authors’ adherence to all the PLOS ONE policies on sharing data and materials.

Figures

Figure 1
Figure 1. Range of mobility during day and night.
This is quantified by the mean daily dwell-time that an average individual in Ivory Coast (right plot) and Portugal (left plot) spends in each of his/her ranked places in the set of non-overlapping frequented places, plotted on a log-log scale, during the day (solid red lines with crosses) and during the night (dashed blue lines with open circles). While the daytime curves follow roughly Zipf’s law, the nighttime curves show a distinct sigmoidal behavior.
Figure 2
Figure 2. Distributions of home-work commuting distances, aggregated by countries/cities.
The distributions are plotted for Ivory Coast (blue solid line with closed dots), Portugal (red dashed line with x’s), Saudi Arabia (green solid line with open circles), Boston (black solid line with open diamonds), and Milan (cyan dashed line with open triangles). The inset plot is the same plot, reproduced on a log-log scale to show long-tail behaviors of the distributions. The same plots, as cumulative density functions, are shown in Figure S4 for comparative purposes.
Figure 3
Figure 3. Distributions of commute timing.
The timing of morning (a, c) and evening commutes (b, d) for Ivory Coast (a, b) and Portugal (c, d), for individuals binned by their commute distances: <2.5 km (blue solid line), 2.5–5 km (red dashed line), 5–10 km (green dash-dotted line), 10–20 km (black solid line), and >20 km (cyan dashed line). The individual’s commute times in the morning and evening are estimated, respectively, by the time of the last call from home in the morning, and by the time of the first call from home in the evening. Fig. S3 shows the sample fit of such distribution to a Gaussian distribution.
Figure 4
Figure 4. Peak commute times as a function of commute distance.
The peak times of morning (a, c) and evening (b, d) commutes for Ivory Coast (blue solid line with open circles) and Portugal (red dashed line with closed dots), as a function of commute distance. There are two methods of calculating this peak time: the median time (a, b), and the fitted Gaussian mean time (c, d). Note the stronger distance-dependent behaviors in the morning. The individual’s commute times in the morning and evening are estimated, respectively, by the time of the last call from home in the morning, and by the time of the first call from home in the evening. Fig. S3 shows the procedure whereby Gaussian distributions are fitted to the distributions plotted in Fig. 3 in order to produce the peak commute time values. The statistics from the Spearman’s rank correlation tests on these relationships are summarized in Table 2.
Figure 5
Figure 5. Mean commute times as a function of commute distance.
The means are for the morning (a) and evening (b) for Ivory Coast (blue solid line with closed dots), Portugal (red dashed line with x’s), Saudi Arabia (green solid line with open circles), Boston (black solid line with open diamonds), and Milan (cyan dashed line with open triangles). While Ivory Coast, Portugal, and Boston consist of mobile phone datasets that cover aggregate commute patterns, the Milan data are GPS traces, which provide a comparative insight into car-only commute patterns.
Figure 6
Figure 6. Probability density functions of commute times.
The first column of figures shows the probability density functions of morning commute times based on mobile phone signaling data, in Ivory Coast (a–b), Portugal (c–d), Saudi Arabia (e–f), Boston (g–h), and Milan (i–j), for individuals binned by their commute distances: <5 km (blue solid line), 5–10 km (red dashed line), 10–20 km (green dash-dotted line), 20–40 km (black solid line), and 40–80 km (magenta dashed line). The inset plots show the cumulative distribution function of the same quantities. The second column of figures shows the probability density functions of evening commute times the respective regions.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Brockmann D, Hufnagel L, Giesel T (2006) The scaling laws of human travel. Nature 439: 462–465. - PubMed
    1. Liang X, Zheng X, Lv W, Zhu T, Xu K (2012) The scaling of human mobility by taxis is exponential. Physica A 391: 2135–2144.
    1. Noulas A, Scellato S, Lambiotte R, Massimiliano P, Mascolo C (2012) A tale of many cities: universal patterns in human urban mobility. PLoS ONE 7: e37027. - PMC - PubMed
    1. Rhee I, Shin M, Hong S, Lee K, Kim SJ, et al. (2011) On the Levy-walk nature of human mobility. IEEE Transactions on Networking 19: 630–643.
    1. Jia T, Jiang B, Carling K, Bolin M, Ban Y (2012) An empirical study on human mobility and its agent-based modeling. Journal of Statistical Mechanics 2012: P11024.

Publication types

Grants and funding

Audi, Volkswagen, BBVA, The Coca Cola Company, Ericsson, Ferrovial, GE provided funding for this study. There are no patents, products in development or marketed products to declare. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

LinkOut - more resources