Joseph Ferreira

A review of urban computing for mobile phone traces: current methods, challenges and opportunities

In this work, we present three classes of methods to extract information from triangulated mobile phone signals, and describe applications with different goals in spatiotemporal analysis and urban modeling. Our first challenge is to relate extracted information from phone records (i.e., a set of time-stamped coordinates estimated from signal strengths) with destinations by each of the million anonymous users. By demonstrating a method that converts phone signals into small grid cell destinations, we present a framework that bridges triangulated mobile phone data with previously established findings obtained from data at more coarse-grained resolutions (such as at the cell tower or census tract levels). In particular, this method allows us to relate daily mobility networks, called motifs here, with trip chains extracted from travel diary surveys. Compared with existing travel demand models mainly relying on expensive and less-frequent travel survey data, this method represents an advantage for applying ubiquitous mobile phone data to urban and transportation modeling applications. Second, we present a method that takes advantage of the high spatial resolution of the triangulated phone data to infer trip purposes by examining semantic-enriched land uses surrounding destinations in individuals motifs. In the final section, we discuss a portable computational architecture that allows us to manage and analyze mobile phone data in geospatial databases, and to map mobile phone trips onto spatial networks such that further analysis about flows and network performances can be done. The combination of these three methods demonstrate the state-of-the-art algorithms that can be adapted to triangulated mobile phone data for the context of urban computing and modeling applications.

Clustering daily patterns of human activities in the city

Data mining and statistical learning techniques are powerful analysis tools yet to be incorporated in the domain of urban studies and transportation research. In this work, we analyze an activity-based travel survey conducted in the Chicago metropolitan area over a demographic representative sample of its population. Detailed data on activities by time of day were collected from more than 30,000 individuals (and 10,552 households) who participated in a 1-day or 2-day survey implemented from January 2007 to February 2008. We examine this large-scale data in order to explore three critical issues: (1) the inherent daily activity structure of individuals in a metropolitan area, (2) the variation of individual daily activities—how they grow and fade over time, and (3) clusters of individual behaviors and the revelation of their related socio-demographic information. We find that the population can be clustered into 8 and 7 representative groups according to their activities during weekdays and weekends, respectively. Our results enrich the traditional divisions consisting of only three groups (workers, students and non-workers) and provide clusters based on activities of different time of day. The generated clusters combined with social demographic information provide a new perspective for urban and transportation planning as well as for emergency response and spreading dynamics, by addressing when, where, and how individuals interact with places in metropolitan areas.

TRAVEL BEHAVIOR AT THE HOUSEHOLD LEVEL: UNDERSTANDING LINKAGES WITH RESIDENTIAL CHOICE

Abstract Previous work with data from the Boston Metropolitan Area has suggested that land use characteristics can have measurable impacts on travel behavior such as trip linking and mode choice at the individual level. However, trip planning, especially in households with children or more than one worker, is quite possibly done at the household level. In this paper, we begin to understand the travel behavior choices of households and understand the relationship of these choices with socio-economic characteristics as well as spatial characteristics of the places where the household resides, works and travels through. The results of preliminary models estimated indicate that the travel behavior of a household is indeed related to the households residential location. The models estimated are not for the purposes of travel demand forecasting as in the case of the household based Stockholm models. The results do indicate if land use, network and accessibility characteristics also affect household trip linking and mode choice and their relationship to residential choice. Thus, one can begin to determine whether planners can make a difference through the implementation of the ideas of neo-traditional theories in local level planning. These models should provide a starting point for further exploration of the land use and transportation linkages explored from the point of view of the more realistic unit of the household.

Activity-Based Human Mobility Patterns Inferred from Mobile Phone Data: A Case Study of Singapore

In this study, with Singapore as an example, we demonstrate how we can use mobile phone call detail record (CDR) data, which contains millions of anonymous users, to extract individual mobility networks comparable to the activity-based approach. Such an approach is widely used in the transportation planning practice to develop urban micro simulations of individual daily activities and travel; yet it depends highly on detailed travel survey data to capture individual activity-based behavior. We provide an innovative data mining framework that synthesizes the state-of-the-art techniques in extracting mobility patterns from raw mobile phone CDR data, and design a pipeline that can translate the massive and passive mobile phone records to meaningful spatial human mobility patterns readily interpretable for urban and transportation planning purposes. With growing ubiquitous mobile sensing, and shrinking labor and fiscal resources in the public sector globally, the method presented in this research can be used as a low-cost alternative for transportation and planning agencies to understand the human activity patterns in cities, and provide targeted plans for future sustainable development.

Choices Versus Choice Sets: A Commuting Spectrum Method for Representing Job--Housing Possibilities

Conclusions in empirical studies of commuting and urban spatial structure depend on the selection of measures for the job-housing relationship. In order to help to develop urban growth strategies based on coherent empirical results, this paper presents a ‘commuting spectrum’ approach as an alternative to existing job–housing relationship measures. With this approach, two hypothetical and extreme commuting possibilities are conceptualized as measures for job–housing relationship and location-choice sets. Simulation in a stylized region and empirical results for Boston and Atlanta indicate that the proposed method can track local and regional aspects of job – housing relationship changes. The revealed association between commuting length and the job–housing relationship is consistent from the perspectives of neighborhood-level comparison, multiyear comparison, and interregion comparison.

Residential Property Values and the Built Environment: Empirical Study in the Boston, Massachusetts, Metropolitan Area

By taking advantage of two recent data sets with exceptional spatial detail, this research is a comprehensive and spatially disaggregate study of the relationship between the built environment and residential property values in the Boston, Massachusetts, metropolitan area. The study computes 27 built environment variables at a 250 m × 250 m grid cell level, uses factor analysis to extract five built environment factors to mitigate multicollinearity, and integrates built environment factors into hedonic price models. Spatial regression techniques are applied to correct spatial autocorrelation. Residential property values are found to be positively associated with accessibility to transit and jobs, connectivity, and walkability and negatively related to auto dominance. Built environment effects depend on neighborhood characteristics.

Planning support systems and smart cities

Since natural disasters frequently happen all over the world, we must make effective preparations for such disasters. As the implementation of sophisticated computerization expands, the society now benefits from ubiquitous network and cloud computing. Consequently, we can utilize a variety of information systems effectively for disaster reduction measures. Based on the experiences of natural disasters, among a variety of information systems, the roles of GIS (Geographic Information Systems) and social media are considered important for collection and transmission of disaster information. Against the above-mentioned backdrop, the present study aims to classify disaster risk management for natural disasters into three stages—normal times, disaster outbreak times, and times of recovery and reconstruction—to introduce the results of development and operation of social media GIS during each of these three stages. The social media GIS targeted residents who were more than 18 years old in the Tama region of Tokyo metropolis and the neighboring area in Japan for two months. Subsequently, the systems were evaluated based on the results of an online questionnaire survey to users, access surveys using log data during operation of the systems, and an analysis of the submitted information. Based on the results of the evaluation, measures for improvement of the development and operation of social media GIS can be summarized into three areas regarding (a) participation of various users and partnership with local communities, (b) usability, and (c) long-term actual operation. K. Yamamoto (&) Graduate School of Information Systems, University of Electro-Communications, 1-5-1 Chofugaoka, Chofu, Tokyo 182-8585, Japan e-mail: k-yamamoto@is.uec.ac.jp

Synthetic Population Generation at Disaggregated Spatial Scales for Land Use and Transportation Microsimulation

The execution of agent-based microsimulation requires an initial set of agents with detailed socioeconomic and demographic attributes to support subsequent behavioral and market models. Data limitations and privacy reasons often restrict the scope and detail with which a synthetic population can be generated by the traditional population synthesis approach. For the accommodation of the growing requirement of microsimulation on spatial resolution and variety, considering new data sources that overcome the data limitations and support population synthesis at more disaggregated levels is necessary. This paper presents a two-stage population synthesis approach not only to improve the accuracy of population generation with imperfect microdata and marginal data, but also to use additional data sets when the spatial details of the synthetic population are interpolated. A general iterative proportional fitting (IPF) method is used in the first stage to estimate the joint distribution of household and individual characteristics under multiple levels of constraints. Additional building information is collected from multiple sources and used to estimate spatial patterns of housing and household characteristics that are then preserved through a second IPF procedure. Preliminary tests of the proposed two-stage IPF-based approach with Singapore data show that the method yields better fitted population realizations at more fine-grained levels than do traditional one-step population synthesis methods.

Pedestrian environments, transit path choice, and transfer penalties : understanding land-use impacts on transit travel

This paper investigates the impact of pedestrian environments on walking behavior, and the related choice of travel path for transit riders. Activity logs from trip surveys combined with transit-route and land-use information are used to fit discrete-choice models of how riders choose among multiple paths to downtown destinations. The work illustrates (1) how the quality of pedestrian environments along transit egress paths affects transfers inside a transit system, and (2) how the impedance of transferring affects egress walking path choices. The use of GIS techniques for path-based spatial analysis is key to understanding the impact of pedestrian environments on walking behavior at the street level. The results show that desirable pedestrian environments encourage transit riders to choose paths that are ‘friendlier’, even if they involve more walking after leaving transit. Policy implications for land-use planning and transit service planning are discussed.

Inferring individual daily activities from mobile phone traces: A Boston example

Understanding individual daily activity patterns is essential for travel demand management and urban planning. This research introduces a new method to infer individuals’ activities from their mobile phone traces. Using Metro Boston as an example, we develop an activity detection model with travel diary surveys to reveal the common laws governing individuals’ activity participation, and apply the modeling results to mobile phone traces to extract the embedded activity information. The proposed approach enables us to spatially and temporally quantify, visualize, and examine urban activity landscapes in a metropolitan area and provides real-time decision support for the city. This study also demonstrates the potential value of combining new “big data” such as mobile phone traces and traditional travel surveys to improve transportation planning and urban planning and management.