Universal predictability of mobility patterns in cities

Abstract

Despite the long history of modelling human mobility, we continue to lack a highly accurate approach with low data requirements for predicting mobility patterns in cities. Here, we present a population-weighted opportunities model without any adjustable parameters to capture the underlying driving force accounting for human mobility patterns at the city scale. We use various mobility data collected from a number of cities with different characteristics to demonstrate the predictive power of our model. We find that insofar as the spatial distribution of population is available, our model offers universal prediction of mobility patterns in good agreement with real observations, including distance distribution, destination travel constraints and flux. By contrast, the models that succeed in modelling mobility patterns in countries are not applicable in cities, which suggests that there is a diversity of human mobility at different spatial scales. Our model has potential applications in many fields relevant to mobility behaviour in cities, without relying on previous mobility measurements.

Intra-Urban Movement Flow Estimation Using Location Based Social Networking Data

“…To evaluate the performance of the models we investigated the travel distance distribution by three models based on ground truth data. Travel distance distribution is an important statistical property to capture human mobility behaviours (Yan et al, 2014). As shown in Figure (2) it can be found that the distribution of travel distance estimated by the rank-based model has a good accordance with the ground truth data compared to the other models.…”

“…Since the 1940s, various models such as gravity model (Zipf, 1946), intervening opportunity model (Stouffer, 1940), radiation model (Simini et al, 2012), rank-based model (Noulas et al, 2012) and population-weighted opportunities (PWO) model (Yan et al, 2014) have been proposed for estimation of OD matrix. In this study, the three following models have been utilized to compare their performance for LBSN data analysis:…”

“…The Sørensen similarity index is a statistic tool to identify the similarity between two samples (Yan et al, 2014). In this paper, we were used the index to measure the degree of agreement between reproduced travel matrices and empirical observations.…”

“…Finally, the "population-weighted opportunities model" (Yan et al, 2014) assumes that the attraction of a destination is inversely proportional to the population Sji in the circle centered at the destination with radius rij (the distance between the origin i and destination j) minus a finite-size correction 1/M:…”

A primer for working with the Spatial Interaction modeling (SpInt) module in the python spatial analysis library (PySAL)

“…One final fit statistic, a modified Sorensen similarity index (SSI), is included within the SpInt module because it has become increasingly popular in some spatial interaction literature that deals with non-parametric models (Lenormand et al 2012, Masucci et al 2012, Yan et al 2013. Using the same symbol definition from the SRMSE, the SSI is defined as,…”