Cell Phone Location Data for Travel Behavior Analysis (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

5 This report presents cell phone locational data and its use in understanding travel behavior, evaluates the extent to which cell phone data can be used to accurately reflect daily travel, and offers guidelines for planners using these data to understand and model travel behavior. To provide the proper context, the research team compared and contrasted traditional survey data and models with U.S. Census data on regional travel and with cell phone–derived estimates for regional travel in Boston, Massachusetts. The report can be read in accordance with the background and interests of the reader. The overall value of the data and potential uses are presented at the beginning and the end of the report. Each chapter begins with a roadmap and ends with conclusions highlighting the key points. Chapters 2 and 3 present a policy discussion and the view of cell phone data from a planner’s perspective. This discussion provides background on how available cell phone data and the ongoing research are likely to influence planning for transportation projects in the near future: • Chapter 2 sets the stage for the research and analyses presented, the nature and potential value of cell phone data, and the recommendations that are part of the practitioner guidelines. The research team discusses the need to evaluate the strengths and weaknesses of traditional data and models before focusing on this new stream of data and its applicability to the development and application of a travel demand model. The research team poses questions of interest, dis- cusses current practices regarding survey data and traditional or advanced models, and then introduces cell phone data as a disruptive technology. The chapter concludes with a discussion of the challenge of using “big data” for different aspects of the traditional modeling stream. • Chapter 3 provides an in-depth look at cell phone data and compares it with the needs that transportation planners have to address. The chapter starts with a definition of call detail records (CDRs) that are at the core of most available industry products, discusses how traces from cell phone data are analyzed to provide locations and to infer activities, and focuses on the assumptions made to draw inferences. How the dynamic nature of cell phone data as technology continues to evolve is also noted. The chapter then focuses on the need to measure travel today and to forecast travel in the future with a discussion of how cell phone data can augment or validate existing models and tools. The research team comments on the need for a behavioral mechanism to organize these data. Then the views of transportation planners on cell phone data and the potential value and uses of these data are summarized. The current marketplace for cell phone data products, the planners’ need to understand the underlying assumptions, and the types of questions practitioners are likely to ask when using these products are outlined. The chapter concludes with a presentation of how travel measures produced by traditional models and survey data will be compared with measures of travel that were developed on the basis of the analysis of cell phone data. C H A P T E R 1 Roadmap to the Report

6 Cell Phone Location Data for Travel Behavior Analysis Chapters 4 through 8 provide an in-depth technical discussion of cell phone data, infer- ence of locations and activity types, development of origin–destination (O-D) matrices, and comparisons with survey data to evaluate the robustness of the methods and results. These chapters provide a unique glimpse into the “black box” where cell phone traces are translated into travel patterns. The report also presents a case study from the Boston area that provides a unique perspec- tive on how cell phone data are analyzed to develop outputs consistent with traditional model results and presents the strengths and weaknesses of such data. The reader interested in an in- depth literature review, the evolution of research in this area, and practical multiway compari- sons of cell phone–derived measures of travel with traditional model output will find Chapters 4 through 8 of particular interest. • Chapter 4 provides an overview of the raw data and describes the range of spatial and temporal resolutions of cell phone data. It demonstrates the massive and passive nature of raw cell phone data and explains in detail their spatial and temporal characteristics. The analysis relies on CDR data from 2 million cell phones collected over 2 months in the Boston region. • Chapter 5 presents methods to extract stay locations where individuals conduct daily activities as anchor points. The spatial and temporal patterns of extracted stay locations for the Boston region are shown. • Chapter 6 focuses on methods, results, and validation to label activity types for “home,” “work,” and “other” stay locations. These activities provide the cornerstone for the estimation of O-D trip matrices. Also discussed are the factors developed to expand the data from cell phone users to the metropolitan population. The expanded home and work activity estimates are compared with journey-to-work travel data. • Chapter 7 presents methods for estimating O-D matrices by purpose and by time of day on the basis of identified activity types and expansion factors from cell phone data and Census population data. The results discussed are analogous to outputs of travel demand models and include estimates of trip generation and trip distribution. • Chapter 8 compares the O-D flows estimated from raw cell phone data for 2 months in 2010 with other sources of travel estimates. These sources include the 2009 National Household Travel Survey; the 2011 Massachusetts Travel Survey; the 1991 Boston Household Travel Survey; the 2007 and 2010 versions of the regional Boston model maintained by the Central Transportation Planning Staff; and a 2015 third-party proprietary data set purchased from a CDR data vendor. Concept of the Black Box The complexity of travel demand models can lead the public and decision makers to think of models as a black box whose internal structure and underlying assumptions are not known or clearly understood. The modeling community has made efforts over the years to better document, explain, and communicate data, methodologies, and assumptions to decision makers and to the public. A similar level of transparency is critical when new methodologies and assumptions are used to analyze cell phone data to support planning decisions and build travel demand models.

Roadmap to the Report 7 Chapter 9 provides guidelines for planners that summarize key issues that analysts con- sider and the potential uses of cell phone data. The discussion combines the policy context with the technical analyses to provide guidance to transportation planners and modelers who evaluate different sources of data and models. The chapter also discusses recent trends in cell phone research, including data collection technologies that combine traditional diary-based methods with the power of locational data reflecting cell phone use. Chapter 9 should be of interest to a broad spectrum of transportation practitioners. The report includes a glossary of terms for this new subject area and references current at the time of writing (late 2016).