Since there are no currently active contests, we have switched Climate CoLab to read-only mode.
Learn more at https://climatecolab.org/page/readonly.
Skip navigation
Share via:

Pitch

Using new mobility data, we can know how people move and consume energy, a major cause of climate change. A revolution in urban planning.


Description

THERMODYNAMICS OF HUMAN MOBILITY

GALEN J. WILKERSON 

GJWILKERSON@GMAIL.COM

Problem statement

As many results from Complex Systems methods and data are so new, very little work has been done to estimate actual energy thermodynamics through the physics of human mobility.

There has recently been a great increase in the availability of data - particularly human mobility using mobile phone data, ’tweets’ and Google searches. This data enables modern researchers in the sustainability arena and related fields, such as planning and economics, to understand human mobility at an unprecedented level of detail, along with its consequent large-scale energy consumption [17, 18].

Little question remains of the link between urban form and energy consumption [4, 11, 19], or of the link between large-scale energy consumption and climate change [5]. We can now begin to understand macro-scale urban efficiency in the context of actual mobility, not just urban form, and why some cities are efficient in their populations’ usage of energy, while others are not. Given the exigent nature of the environmental and resource crisis these results are very important.

Methodology

Detailed anonymous information can be extracted from mobile phone [9], Google query location [14], geotagged Twitter data [7], or other sources [13]. This data enables us to estimate several characteristics:

  1. (1)  Detailed location and time information for users can be found. This can be estimated from the location of phone calls a person makes or receives in a particular time window.

  2. (2)  The geographic centroid of a particular user’s locations can be calculated.

  3. (3)  A heuristic can be used to estimate the household location of a user - it can be estimated from the

    phone calls a person makes or receives between late evening and early morning.

Upon estimating (1-3) above, the tools of Complex Systems, partly originating in Statistical Physics, may be applied [2, 9, 18] to extract fundamental mathematical relationships describing how people move in a city, perhaps even developing a Hamiltonian formalism for regional mobility patterns.

Date: May 2, 2012.

(a) Kinetic Energy Model (b) Potential Energy Model (c) Simulated Heat Map
Figure 1. Mobility Energy Models: (a) the Kinetic Energy Model, Ek ? v ̄i2, for individuals i, where v ̄ ∆d . (b), the Potential Energy Model, U ∆d or ∆d2 (to be studied). (c) simulates ∆t a heat map (source: walkscore.com).

 

 

 

Key Research Questions

From the above information (1-3), we may begin to answer several very basic questions. Respectively:

(1) What are the locations a person visits over a particular time period? (2) What is the centroid (i.e. geographic center) of these locations?
(3) How far does the person live from this centroid?

Additionally, only knowing (1) above, we may ask, what is the thermodynamic energy, entropy, and temperature of mobility for a particular city? (Figure 1a.) Geographic heat maps can be made showing average values for these measures in a particular location. (Figure 1c.) Cities can be evaluated and compared.

Also, when we apply (2) and (3) above, we might ask practical questions about energy consumption. Applying the poly-nuclear urban model [10], if we assume an ideal city where everyone lives at the centroid of their geographic activity, how far, on average, do these real people live from that centroid? That gives us some macro-scale measure of potential energy. (Figure 1b.) Again, heat maps can be made showing this geographically.

We can then estimate a mean-field surplus energy consumption from the above measurements. In other words, on average, how far is fuel consumption from the ideal, where everyone lives at the centroid of their activity?

Alternatively, a similar analysis could be carried out vis-a-vis the concentric urban model [12, 20], esti- mating potential energy as distance of household and/or mobility from the city center.

We may also begin to make a connection between social ties and energy through mobility [6, 21].

 

 

Expected Contribution to Knowledge

I expect that we will begin to understand cities and their energy budget at a level of detail that is unprecedented, and which allows us to characterize them in a way that is very useful to planners and policymakers. We may also begin to extract fundamental physical properties of human mobility, especially in the context of energy consumption. This extends the work in mobility of Bettencourt, Gonzalez, Barabasi, Brockmann, Helbing, and others into the arena of energy and sustainability [1, 2, 3, 6, 8, 9, 15, 16, 18].

The proposed work will let us:

(1) Gain very practical detailed information about the general mobility patterns of cities.
(2) Learn about efficiency of cities. This information can be represented graphically as a
heat map.(3) Develop a theory of mobility and efficiency in the context of thermodynamic concepts of entropy, energy, and temperature.
(4) Connect these efficiency measures to actual large-scale fossil fuel consumption.

Time Frame and resources

It is expected that a preliminary investigation into these measures would take at least 3 months.  Data processing, debugging, analysis, and evaluation all should be done with due deliberation and care. This project has the potential to continue as PhD research.

To obtain optimum results, I hope to collaborate with a Statistical Physicist who has a strong interest in sustainability to develop these ideas and methods.

Mobile phone datasets containing time and location information, along with anonymized user IDs, are needed for this analysis. Alternatively, geotagged Twitter ’tweets’ can be downloaded using the API or from various sources. Other useable geotagged datasets are Gowalla or Brightkite at the Stanford SNAP database.

I expect that I will process data in C++ or Python, both of which I have used extensively in the past. I also have experience with Geographic Information Systems and Statistical Physics methodologies.

Ideally, a server having Linux or Unix and a reasonably large memory (5-10GB) and disk space (1TB) would be available. Certainly, productive work can be done on a more limited server. Preliminary development could be done on a laptop or workstation.

References

  1. James P. Bagrow and Yu-Ru Lin, Spatiotemporal features of human mobility, (2012).

  2. Lu ́ıs M. A. Bettencourt, Jos ́e Lobo, Dirk Helbing, Christian Ku ̈hnert, and Geoffrey B. West, Growth, innovation, scaling,

    and the pace of life in cities, Proceedings of the National Academy of Sciences 104 (2007), no. 17, 7301–7306.

  3. D. Brockmann, L. Hufnagel, and T. Geisel, The scaling laws of human travel, Nature 439 (2006), no. 7075, 462–465.

  4. Roberto Camagni, Maria Cristina Gibelli, and Paolo Rigamonti, Urban mobility and urban form: the social and environ-

    mental costs of different patterns of urban expansion, Ecological Economics 40 (2002), no. 2, 199 – 216.

  5. Lee Chapman, Transport and climate change: a review, Journal of Transport Geography 15 (2007), no. 5, 354 – 367.

RESEARCH PROPOSAL - THERMODYNAMICS OF HUMAN MOBILITY 3

  1. Eunjoon Cho, Seth A. Myers, and Jure Leskovec, Friendship and mobility: user movement in location-based social networks, Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining (New York, NY, USA), KDD ’11, ACM, 2011, pp. 1082–1090.

  2. Jacob Eisenstein, Brendan O’Connor, Noah A. Smith, and Eric P. Xing, A latent variable model for geographic lexical variation, Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing (Stroudsburg, PA, USA), EMNLP ’10, Association for Computational Linguistics, 2010, pp. 1277–1287.

  3. F. Simini et. al., A universal model for mobility and migration patterns, Nature 484 (2012), no. 96.

  4. Marta C. Gonzalez, Cesar A. Hidalgo, and Albert-Laszlo Barabasi, Understanding individual human mobility patterns,

    Nature 453 (2008), no. 7196, 779–782.

  5. Chauncy D. Harris and Edward L. Ullman, The nature of cities, The ANNALS of the American Academy of Political and

    Social Science 242 (1945), no. 1, 7–17.

  6. Peter W. G. Newman and Jeffrey R. Kenworthy, Gasoline consumption and cities, Journal of the American Planning

    Association 55 (1989), no. 1, 24–37.

  7. Ernest W. Burgess Park, Robert and Roderick D. McKenzie, The city, University of Chicago Press, 1925.

  8. A. Pentland, Funf open sensing framework, "http://funf.media.mit.edu/".

  9. Deepak Ravichandran D. Sivakumar Shumeet Baluja Rohan Seth, Michele Covell, A tale of two (similar) cities: Inferring

    city similarity through geo-spatial query log analysis, Proceedings of the International Conference on Knowledge Discovery

    and Information Retrieval (2011).

  10. Chaoming Song, Tal Koren, Pu Wang, and Albert-Laszlo Barabasi, Modelling the scaling properties of human mobility, Nat

    Phys 6 (2010), no. 10, 818–823.

  11. Chaoming Song, Zehui Qu, Nicholas Blumm, and Albert-La ́szlo ́ Baraba ́si, Limits of predictability in human mobility, Science

    327 (2010), no. 5968, 1018–1021.

  12. Anthony M Townsend, Life in the real-time city: Mobile telephones and urban metabolism, Group 7 (2000), no. 212, 85–104.

  13. TRANTOPOULOS;Konstantinos, SCHLAPFER;Markus, and HELBING;Dirk, Toward sustainability of complex urban sys-

    tems through techno-social reality mining, 2011, p. 2.

  14. A. Troy, The very hungry city: Urban energy efficiency and the economic fate of cities, Yale University Press, 2012.

  15. J.H. Von Thuenen, Der isolierte staat in beziehung auf landwirtschaft und nationaloekonomie, G Fischer, 1910.

  16. Dashun Wang, Dino Pedreschi, Chaoming Song, Fosca Giannotti, and Albert-Laszlo Barabasi, Human mobility, social ties,

    and link prediction, Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining (New York, NY, USA), KDD ’11, ACM, 2011, pp. 1100–1108. 

Summary


Category of the action

Reducing emissions from transportation


What actions do you propose?

THERMODYNAMICS OF HUMAN MOBILITY

GALEN J. WILKERSON 

GJWILKERSON@GMAIL.COM

Problem statement

As many results from Complex Systems methods and data are so new, very little work has been done to estimate actual energy thermodynamics through the physics of human mobility.

There has recently been a great increase in the availability of data - particularly human mobility using mobile phone data, ’tweets’ and Google searches. This data enables modern researchers in the sustainability arena and related fields, such as planning and economics, to understand human mobility at an unprecedented level of detail, along with its consequent large-scale energy consumption [17, 18].

Little question remains of the link between urban form and energy consumption [4, 11, 19], or of the link between large-scale energy consumption and climate change [5]. We can now begin to understand macro-scale urban efficiency in the context of actual mobility, not just urban form, and why some cities are efficient in their populations’ usage of energy, while others are not. Given the exigent nature of the environmental and resource crisis these results are very important.

Methodology

Detailed anonymous information can be extracted from mobile phone [9], Google query location [14], geotagged Twitter data [7], or other sources [13]. This data enables us to estimate several characteristics:

  1. (1)  Detailed location and time information for users can be found. This can be estimated from the location of phone calls a person makes or receives in a particular time window.

  2. (2)  The geographic centroid of a particular user’s locations can be calculated.

  3. (3)  A heuristic can be used to estimate the household location of a user - it can be estimated from the

    phone calls a person makes or receives between late evening and early morning.

Upon estimating (1-3) above, the tools of Complex Systems, partly originating in Statistical Physics, may be applied [2, 9, 18] to extract fundamental mathematical relationships describing how people move in a city, perhaps even developing a Hamiltonian formalism for regional mobility patterns.

Date: May 2, 2012.

(a) Kinetic Energy Model (b) Potential Energy Model (c) Simulated Heat Map
Figure 1. Mobility Energy Models: (a) the Kinetic Energy Model, Ek ? v ̄i2, for individuals i, where v ̄ ∆d . (b), the Potential Energy Model, U ∆d or ∆d2 (to be studied). (c) simulates ∆t a heat map (source: walkscore.com).

 

 

 

Key Research Questions

From the above information (1-3), we may begin to answer several very basic questions. Respectively:

(1) What are the locations a person visits over a particular time period? (2) What is the centroid (i.e. geographic center) of these locations?
(3) How far does the person live from this centroid?

Additionally, only knowing (1) above, we may ask, what is the thermodynamic energy, entropy, and temperature of mobility for a particular city? (Figure 1a.) Geographic heat maps can be made showing average values for these measures in a particular location. (Figure 1c.) Cities can be evaluated and compared.

Also, when we apply (2) and (3) above, we might ask practical questions about energy consumption. Applying the poly-nuclear urban model [10], if we assume an ideal city where everyone lives at the centroid of their geographic activity, how far, on average, do these real people live from that centroid? That gives us some macro-scale measure of potential energy. (Figure 1b.) Again, heat maps can be made showing this geographically.

We can then estimate a mean-field surplus energy consumption from the above measurements. In other words, on average, how far is fuel consumption from the ideal, where everyone lives at the centroid of their activity?

Alternatively, a similar analysis could be carried out vis-a-vis the concentric urban model [12, 20], esti- mating potential energy as distance of household and/or mobility from the city center.

We may also begin to make a connection between social ties and energy through mobility [6, 21].

 

 

Expected Contribution to Knowledge

I expect that we will begin to understand cities and their energy budget at a level of detail that is unprecedented, and which allows us to characterize them in a way that is very useful to planners and policymakers. We may also begin to extract fundamental physical properties of human mobility, especially in the context of energy consumption. This extends the work in mobility of Bettencourt, Gonzalez, Barabasi, Brockmann, Helbing, and others into the arena of energy and sustainability [1, 2, 3, 6, 8, 9, 15, 16, 18].

The proposed work will let us:

(1) Gain very practical detailed information about the general mobility patterns of cities.
(2) Learn about efficiency of cities. This information can be represented graphically as a
heat map.(3) Develop a theory of mobility and efficiency in the context of thermodynamic concepts of entropy, energy, and temperature.
(4) Connect these efficiency measures to actual large-scale fossil fuel consumption.

Time Frame and resources

It is expected that a preliminary investigation into these measures would take at least 3 months.  Data processing, debugging, analysis, and evaluation all should be done with due deliberation and care. This project has the potential to continue as PhD research.

To obtain optimum results, I hope to collaborate with a Statistical Physicist who has a strong interest in sustainability to develop these ideas and methods.

Mobile phone datasets containing time and location information, along with anonymized user IDs, are needed for this analysis. Alternatively, geotagged Twitter ’tweets’ can be downloaded using the API or from various sources. Other useable geotagged datasets are Gowalla or Brightkite at the Stanford SNAP database.

I expect that I will process data in C++ or Python, both of which I have used extensively in the past. I also have experience with Geographic Information Systems and Statistical Physics methodologies.

Ideally, a server having Linux or Unix and a reasonably large memory (5-10GB) and disk space (1TB) would be available. Certainly, productive work can be done on a more limited server. Preliminary development could be done on a laptop or workstation.

References

  1. James P. Bagrow and Yu-Ru Lin, Spatiotemporal features of human mobility, (2012).

  2. Lu ́ıs M. A. Bettencourt, Jos ́e Lobo, Dirk Helbing, Christian Ku ̈hnert, and Geoffrey B. West, Growth, innovation, scaling,

    and the pace of life in cities, Proceedings of the National Academy of Sciences 104 (2007), no. 17, 7301–7306.

  3. D. Brockmann, L. Hufnagel, and T. Geisel, The scaling laws of human travel, Nature 439 (2006), no. 7075, 462–465.

  4. Roberto Camagni, Maria Cristina Gibelli, and Paolo Rigamonti, Urban mobility and urban form: the social and environ-

    mental costs of different patterns of urban expansion, Ecological Economics 40 (2002), no. 2, 199 – 216.

  5. Lee Chapman, Transport and climate change: a review, Journal of Transport Geography 15 (2007), no. 5, 354 – 367.

RESEARCH PROPOSAL - THERMODYNAMICS OF HUMAN MOBILITY 3

  1. Eunjoon Cho, Seth A. Myers, and Jure Leskovec, Friendship and mobility: user movement in location-based social networks, Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining (New York, NY, USA), KDD ’11, ACM, 2011, pp. 1082–1090.

  2. Jacob Eisenstein, Brendan O’Connor, Noah A. Smith, and Eric P. Xing, A latent variable model for geographic lexical variation, Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing (Stroudsburg, PA, USA), EMNLP ’10, Association for Computational Linguistics, 2010, pp. 1277–1287.

  3. F. Simini et. al., A universal model for mobility and migration patterns, Nature 484 (2012), no. 96.

  4. Marta C. Gonzalez, Cesar A. Hidalgo, and Albert-Laszlo Barabasi, Understanding individual human mobility patterns,

    Nature 453 (2008), no. 7196, 779–782.

  5. Chauncy D. Harris and Edward L. Ullman, The nature of cities, The ANNALS of the American Academy of Political and

    Social Science 242 (1945), no. 1, 7–17.

  6. Peter W. G. Newman and Jeffrey R. Kenworthy, Gasoline consumption and cities, Journal of the American Planning

    Association 55 (1989), no. 1, 24–37.

  7. Ernest W. Burgess Park, Robert and Roderick D. McKenzie, The city, University of Chicago Press, 1925.

  8. A. Pentland, Funf open sensing framework, "http://funf.media.mit.edu/".

  9. Deepak Ravichandran D. Sivakumar Shumeet Baluja Rohan Seth, Michele Covell, A tale of two (similar) cities: Inferring

    city similarity through geo-spatial query log analysis, Proceedings of the International Conference on Knowledge Discovery

    and Information Retrieval (2011).

  10. Chaoming Song, Tal Koren, Pu Wang, and Albert-Laszlo Barabasi, Modelling the scaling properties of human mobility, Nat

    Phys 6 (2010), no. 10, 818–823.

  11. Chaoming Song, Zehui Qu, Nicholas Blumm, and Albert-La ́szlo ́ Baraba ́si, Limits of predictability in human mobility, Science

    327 (2010), no. 5968, 1018–1021.

  12. Anthony M Townsend, Life in the real-time city: Mobile telephones and urban metabolism, Group 7 (2000), no. 212, 85–104.

  13. TRANTOPOULOS;Konstantinos, SCHLAPFER;Markus, and HELBING;Dirk, Toward sustainability of complex urban sys-

    tems through techno-social reality mining, 2011, p. 2.

  14. A. Troy, The very hungry city: Urban energy efficiency and the economic fate of cities, Yale University Press, 2012.

  15. J.H. Von Thuenen, Der isolierte staat in beziehung auf landwirtschaft und nationaloekonomie, G Fischer, 1910.

  16. Dashun Wang, Dino Pedreschi, Chaoming Song, Fosca Giannotti, and Albert-Laszlo Barabasi, Human mobility, social ties,

    and link prediction, Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining (New York, NY, USA), KDD ’11, ACM, 2011, pp. 1100–1108. 


Who will take these actions?


Where will these actions be taken?


How much will emissions be reduced or sequestered vs. business as usual levels?


What are other key benefits?


What are the proposal’s costs?


Time line


Related proposals


References