Estimating internal displacement in Ukraine from high-frequency GPS mobile phone data

Nearly 110 million people are forcibly displaced people worldwide. However, estimating the scale and patterns of internally displaced persons in real time, and developing appropriate policy responses, remain hindered by traditional data streams. They are infrequently updated, costly and slow. Mobile phone location data can overcome these limitations, but only represent a population segment. Drawing on an anonymised large scale, high-frequency dataset of locations from 25 million mobile devices, we propose an approach to leverage mobile phone data and produce population-level estimates of internal displacement.We use this approach to quantify the extent, pace and geographic patterns of internal displacement in Ukraine during the early stages of the Russian invasion in 2022. Our results produce reliable population-level estimates, enabling real-time monitoring of internal displacement at detailed spatio-temporal resolutions. Accurate estimations are crucial to support timely and e ective humanitarian and disaster management responses, prioritising resources where they are most needed.

The Department for Transport (DfT) in the UK is evaluating the process of obtaining traffic statistics such as those collected in the National Travel Survey and the traffic census. Alongside traditional methods, such as surveys and loop data, mobile phone data can be used to support and complete the existing methods. As well as considering the potential of ITS to compile traffic statistics, the DfT is also interested in its applicability to the measurement of road freight (HGV and LGV) activity, particularly origins and destinations data. In 1998, TRL was commissioned by the Highways Agency to conduct research into the feasibility of using mobile phone location data to obtain traffic information, in particular origin-destination (OD) information which would provide the core data required by many transportation simulation models including SATURN and SISTM. The results of the research are described in several documents, including White and Quick (2000) and White and Wells (2002). The research found that it was feasible to obtain OD information from mobile phone location data, as well as other forms of traffic information such as journey times and speeds. Routeing information was also extractable. TRL developed an algorithm to analyse anonymous billing data, kindly provided by BTCellnet, now O/sub 2/. The paper provides an update on the previous work and focuses on the use of mobile phone data for the Department for Transport's requirements.

The advent of big data has aided understanding of the driving forces of human mobility, which is beneficial for many fields, such as mobility prediction, urban planning, and traffic management. However, the data sources used in many studies, such as mobile phone location and geo-tagged social media data, are sparsely sampled in the temporal scale. An individual’s records can be distributed over a few hours a day, or a week, or over just a few hours a month. Thus, the representativeness of sparse mobile phone location data in characterizing human mobility requires analysis before using data to derive human mobility patterns. This paper investigates this important issue through an approach that uses subscriber mobile phone location data collected by a major carrier in Shenzhen, China. A dataset of over 5 million mobile phone subscribers that covers 24 h a day is used as a benchmark to test the representativeness of mobile phone location data on human mobility indicators, such as total travel distance, movement entropy, and radius of gyration. This study divides this dataset by hour, using 2- to 23-h segments to evaluate the representativeness due to the availability of mobile phone location data. The results show that different numbers of hourly segments affect estimations of human mobility indicators and can cause overestimations or underestimations from the individual perspective. On average, the total travel distance and movement entropy tend to be underestimated. The underestimation coefficient results for estimation of total travel distance are approximately linear, declining as the number of time segments increases, and the underestimation coefficient results for estimating movement entropy decline logarithmically as the time segments increase, whereas the radius of gyration tends to be more ambiguous due to the loss of isolated locations. This paper suggests that researchers should carefully interpret results derived from this type of sparse data in the era of big data.

A data fusion approach with mobile phone data for updating travel survey-based mode split estimates

Recent disasters have shown the existence of large variance in recovery trajectories across cities that have experienced similar damage levels. Case studies of such events reveal the high complexity of the recovery process of cities, where inter-city dependencies and intra-city coupling of social and physical systems may affect the outcomes in unforeseen ways. Despite the large implications of understanding the recovery processes of cities after disasters for many domains including critical services, disaster management, and public health, little work have been performed to unravel this complexity. Rather, works are limited to analyzing and modeling cities as independent entities, and have largely neglected the effect that inter-city connectivity may have on the recovery of each city. Large scale mobility data (e.g. mobile phone data, social media data) have enabled us to observe human mobility patterns within and across cities with high spatial and temporal granularity. In this paper, we investigate how inter-city dependencies in both physical as well as social forms contribute to the recovery performances of cities after disasters, through a case study of the population recovery patterns of 78 Puerto Rican counties after Hurricane Maria using mobile phone location data. Various network metrics are used to quantify the types of inter-city dependencies that play an important role for effective post-disaster recovery. We find that inter-city social connectivity, which is measured by pre-disaster mobility patterns, is crucial for quicker recovery after Hurricane Maria. More specifically, counties that had more influx and outflux of people prior to the hurricane, were able to recover faster. Our findings highlight the importance of fostering the social connectivity between cities to prepare effectively for future disasters. This paper introduces a new perspective in the community resilience literature, where we take into account the inter-city dependencies in the recovery process rather than analyzing each community as independent entities.

Urban green space is closely related to the quality of life of residents. However, the traditional approach to its planning often fails to address its actual service capacity and users’ demand. In this study, facilitated by mobile phone location data, more specific features of the spatial distribution of urban residents are identified. Further, population distribution in relation to traffic analysis zones is mapped. On this basis, the two-step floating catchment area method (2SFCA) is adopted in combination with urban green space planning to evaluate the per capita area of green space and its accessibility in practice. Subsequently, classification of per capita area and spatial distribution of green spaces within the study area are obtained; thus, urban districts currently with low accessibility to green areas are identified and can be deemed as key areas for the planning of green areas in the future. The study concludes that mobile phone data can be used to more accurately map the spatial distribution of residents; while, the 2SFCA offers a more comprehensive quantitative measuring of the supply and demand of green spaces. The two combined can be used as an important basis for decision-making in the planning of urban green spaces. Since urban green space can be regarded as a kind of public facility, the methodology of the present study is also believed to be applicable in studies of other types of urban facilities.

In the recent decade, a new concept, urban community life circle (CLC), has been introduced and widely applied to Chinese community planning and public service facilities configuration alongside people-oriented urbanization. How to delineate the CLC has become a core task of urban CLC planning. The traditional way to determine the CLC using administrative boundaries does not fully consider the needs of residents. Recent research on urban CLC delineation is usually based on residential behavior survey using sample surveys or GPS data. However, it is difficult to generalize the sample surveys or GPS surveys for one specific community to that for others, because of the extremely high cost. Due to the ubiquity of the location-based service (LBS) data, i.e., the mobile phone data and points of interest (POI) data, they can serve as a fine-grained and continuous proxy for conducting human daily activity research with easy accessibility and low cost. Mobile phone data can represent the daily travel activities of residents, and POI data can comprehensively describe the physical conditions. In this paper, we propose a method from both the social and physical perspectives to delineate the CLC based on mobile phone and POI data, named DMP for short. The proposed DMP method is applied to Wuhan. We decipher the CLC’s boundary and residents’ travel activity patterns and demonstrate that (1) the CLC is not a regular circle but a non-homogeneous corridor space extending along streets; and (2) adjacent CLCs are found to share some daily facilities. Based on these findings, we propose that CLC planning should be data-based and people-oriented in general. In addition, sufficient space in the overlapping region of the CLCs should be preserved for future planning of public service facilities configuration, given that adjacent CLCs share some daily facilities.

Data collected from mobile phones have potential knowledge to provide background information of users, which are quite valuable to many location-aware applications. In the current research, there is relatively few commercial software or application systems to fully meet the requirements of effectively analyzing these characteristics of urban activity. In the paper, we provide with a complete framework for the applications of urban activity analysis from mobile phone location data, including individual behavioral pattern, comparison among individuals, and group activity. We use actual mobile phone data to implement these functions of discovering pattern and regularity of urban activity to show our rich applications related to mobile phone under our framework.

Estimating freeway traffic measures from mobile phone location data

The COVID-19 pandemic affected many aspects of human mobility and resulted in unprecedented changes in population dynamics, including lifestyle and mobility. Recognizing the effects of the pandemic is crucial to understand changes and mitigate negative impacts. Spatial data on human activity, including mobile phone data, has the potential to provide movement patterns and identify regularly visited locations. Moreover, crowdsourced geospatial information can explain and characterize the regularly visited locations. The analysis of both mobility and routine locations in the same study has seldom been carried out using mobile phone data and linked to the effects of the pandemic. Therefore, in this article we study human mobility patterns within Portugal, using mobile phone and crowdsourced data to compare the population’s mobility and routine locations after the pandemic’s peak. We use clustering algorithms to identify citizens’ stops and routine locations, at an antenna level, during the following months after the pandemic’s first wave and the same period of the following year. Results based on two mobile phone datasets showed a significant difference in mobility in the two periods. Nevertheless, routine locations slightly differ.

Background: Hypertensive disorders of pregnancy (HDPs) are a group of conditions that contribute to maternal morbidity and mortality, disproportionately affecting minority and low-income individuals. Complications from these HDPs may result in poor outcomes for mother and child during pregnancy or post-pregnancy. Disparities in HDPs may be associated with concentrated racial and income inequality in an individual’s activity space, such that the social exposure of places is not only formed by residence but also by daily mobility of the population at large. A novel aspect of women’s residential neighborhood environment can be explored by measuring daytime population mixing. Mobile phone data can be used to estimate social mixing during the daytime for income and racial/ethnic groups by describing the undercurrents of population mobility in neighborhoods. Objective: To estimate the association of daytime racial and income segregation with HDPs among women who gave birth from 2018 to 2019 in metro Atlanta. Methods: Data for women who gave birth in Metro Atlanta in 2018 and 2019 were abstracted from vital statistics records to determine HDP diagnosis and maternal residence by census tract. HDPs were grouped into pre-pregnancy hypertension, gestational hypertension, and eclampsia. Mobile phone location data from CUEBIQ were used to estimate median household income density in each census tract and racial/ethnic density, as a function of where devices moved between the hours of 7am and 10pm. Racial and economic composition of the device was assigned by the user’s estimated home census tract. The Index of Concentration at Extremes (ICE) was calculated as the mixing of mobile devices in each census tract, by race and income. Tracts were categorized as concentrated disadvantage for an ICE score less than -0.3 and concentrated privileged for an ICE score greater than 0.3. Logistic regression models were used to estimate odds of HDPs in census tracts with micro-segregation and income inequality during the daytime, compared to not at extremes. Results: There were 122,482 births included in the maternal cohort, and 8.1% of births occurred among women with a diagnosed HDP. Black women had a higher rate of HDPs compared to White women (9.7% versus 6.9%). Census tracts with higher daytime micro-segregation were associated with higher odds of HDPs compared to tracts not in the extremes (OR = 1.42; 95% CI= 1.35, 1.49). Census tracts with higher daytime income inequality were also associated with higher odds of HDPs (1.26;1.20, 1.33). Conclusions: The odds of maternal HDPs were higher among women in census tracts with greater racial daytime segregation and more income disparity, compared to tracts that were not identified as extremes. Describing and distinguishing social epidemiologic patterns for HDPs by population mobility may provide a better understanding of the social context of who is most at risk.

HomeCirculationVol. 147, No. 24Correction to: 2023 EPI/Lifestyle Abstract 17 Free AccessCorrectionPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessCorrectionPDF/EPUBCorrection to: 2023 EPI/Lifestyle Abstract 17 Originally published12 Jun 2023https://doi.org/10.1161/CIR.0000000000001159Circulation. 2023;147:e721This article corrects the followingAbstract 17: Using Mobile Phone Location Data to Estimate the Association of Daytime Racial and Economic Segregation With Hypertensive Disorders of Pregnancy in Metro AtlantaIn the 2023 EPI Lifestyle abstract by Campbell et al, “Abstract 17: Using Mobile Phone Location Data to Estimate the Association of Daytime Racial and Economic Segregation With Hypertensive Disorders of Pregnancy in Metro Atlanta,” which published on February 28, 2023 (Circulation. 2023;147(suppl):A17. doi: 10.1161/circ.147.suppl_1.17), a correction notice is needed.In the “Background” section, sentence 5, “Anonymized” was added before “mobile phone data can be…”In the “Methods” section, the beginning of sentence 3 read, “Mobile phone location data from CUEBIQ were used to estimate racial/ethnic density…” It has been updated to read, “Aggregated and anonymized mobile-phone data from opted-in devices were used to estimate racial/ethnic density…”In the “Methods” section, sentence 4 read, “Racial and economic composition of the device was assigned by the user’s estimated home census tract.” It has been updated to read, “Racial and economic composition was assigned by the device’s inferred home census tract.”At the end of the “Methods” section, a sentence was added: “At no point was mobility data linked to individual health records.”This correction has been made to the current online version of the abstract, which is available at: https://www.ahajournals.org/doi/10.1161/circ.147.suppl_1.17FootnotesCirculation is available at www.ahajournals.org/journal/circ Previous Back to top Next FiguresReferencesRelatedDetailsRelated articlesAbstract 17: Using Mobile Phone Location Data to Estimate the Association of Daytime Racial and Economic Segregation With Hypertensive Disorders of Pregnancy in Metro AtlantaKatherine Campbell, et al. Circulation. 2023;147:A17-A17 June 13, 2023Vol 147, Issue 24 Advertisement Article InformationMetrics © 2023 American Heart Association, Inc.https://doi.org/10.1161/CIR.0000000000001159PMID: 37307307 Originally publishedJune 12, 2023 PDF download Advertisement

The role of alcohol outlet visits derived from mobile phone location data in enhancing domestic violence prediction at the neighborhood level.

GIScience 2016 Short Paper Proceedings Deriving Hospital Catchment Areas from Mobile Phone Data Bernd Resch 1,2 , Azmat Arif 1 , Gautier Krings 3 , Guillaume Vankeerberghen 3 , Marc Buekenhout 4 University of Salzburg, Department of Geoinformatics – Z_GIS, Schillerstrasse 30, 5020 Salzburg, Austria Email: bernd.resch@sbg.ac.at, azmat.arif@stud.sbg.ac.at Harvard University, Center for Geographical Analysis, 1737 Cambridge Street, Cambridge, MA 02138, USA Real Impact Analytics, 5 Place du Champ de Mars, 1050 Brussels, Belgium Email: {gautier.krings; guillaume.vankeerberghen}@realimpactanalytics.com Email: mbuekenhout@msn.com Abstract Delineating catchment areas of medical facilities is essential for estimating the quality of a health-care system and to maximise the efficiency of health service provision. One critical shortcoming of previous approaches are manifested in their comprehensive assumptions about a hospital’s patients by using census data or gravity models. In contrast, our approach uses anonymised mobile and landline phone data to derive hospital catchment areas. Our goal is not to assess the quality of the health care system, but to identify the geographic areas, in which people actually use a hospital. Thus, our results reveal new insights into the catchment areas of hospitals by minimising assumptions about demographic factors. 1. Introduction and Related Work Adequate provision of health services is a central priority of health professionals and policy makers worldwide. More, the efficiency of health care systems, i.e., the provision of best possible service with minimum resources, is critical to public providers (Fransen et al. 2015). These requirements have led to a number of studies to analyse catchment areas and service quality of medical facilities. Even though geospatial analysis methods have existed for decades, there is still a general lack of studies that have mapped and examined health service catchments in practice, not only from a theoretical viewpoint (Schuurman et al. 2006). Previous approaches for delineating medical catchment areas comprise statistical population-to-provider ratios, gravitational models, travel cost estimation, analysis of the physical distance between hospitals, census-based patient-origin analysis, commuter-based approaches of modelling spatial accessibility (Fransen et al. 2015; Wang and Wheeler 2015), or the two-step floating catchment area method based on the physician-to-population ratio (Luo and Wang 2003). The major drawback of these approaches is that they make far- reaching assumptions about a hospital’s patients by applying census data, travel times or gravity models. Moreover, they do not take heterogeneous activity and mobility patterns into account or only derive them from static census data. Thus, the approach proposed in this paper uses anonymised mobile and landline phone data to delineate hospital catchment areas. Like this, we aim to identify the geographic areas, in which people use a hospital instead of assessing the quality of the health care system per se. Therefore, we analyse calls to and from hospitals in Trinidad and Tobago. This goes beyond just using patient records in that we are able to draw conclusions from a wider range of communication with a hospital (enquiries, arrangement of appointments, follow-up care, visitors, etc.), beyond patients’ hospital stays.

A key issue to understand urban system is to characterize the activity dynamics in a city—when, where, what, and how activities happen in a city. To better understand the urban activity dynamics, city-wide and multiday activity participation sequence data, namely, activity chain as well as suitable spatiotemporal models, are needed. The commonly used household travel survey data in activity analysis suffers from limited sample size and temporal coverage. The emergence of large-scale spatiotemporal data in urban areas, such as mobile phone data, provides a new opportunity to infer urban activities and the underlying dynamics. However, the challenge is the absence of labeled activity information in mobile phone data. Consequently, how to fuse the useful information in household survey data and mobile phone data to build city-wide, multiday, and all-time activity chains becomes an important research question. Moreover, the multidimension structure of the activity data (e.g., location, start time, duration, type) makes the extraction of spatiotemporal activity patterns another difficult problem. In this study, the authors first introduce an activity chain inference model based on tensor decomposition to infer the missing activity labels in large-scale and multiday activity data, and then develop a spatiotemporal event clustering model based on DBSCAN, called STE-DBSCAN, to identify the spatiotemporal activity patterns. The proposed approaches achieved good accuracy and produced patterns with a high level of interpretability.