It's official! We will be presenting at the TRB National Transportation Planning Applications Conference. "Combining passive mobile phone and consumer data: A proof-of-concept study in Atlanta" will detail the work we have undertaken in the past year. The abstract follows:
To date, the collection of comprehensive household travel data has been a challenge due mainly to high costs and the unwillingness of individuals to respond to surveys. Over the past few years, many agencies, consultants, and researchers have investigated other options including passive data collection through GPS, WiFi positioning, Bluetooth, or signal data. However, these types of mobile phone data typically lack information about the individuals making each trip as well as several other important factors like trip mode and purpose.
Meanwhile, consumer data firms have been compiling information about individuals and households for decades, typically selling the inexpensive, up-to-date data to companies wanting to customize marketing campaigns to potential customers. Coincidentally, these readily available data contain the majority of household and individual demographic and socioeconomic fields that are used in travel demand applications. Having said that, consumer data lacks actual trip-making behavior.
By combining these two types of data using statistical, simulation, and data fusion techniques, their respective shortcomings for use in travel demand modeling applications might be overcome. In December 2014, a proof-of-concept study, which combines different types of mobile phone and consumer data to create a synthetic household travel dataset, will be completed in the city of Atlanta. This presentation will report on the findings of the study. A brief description of the types of data used will be provided. The methodology for combining the two data sources into a synthetic household travel dataset will be presented, as will a discussion of the successes and failures of the study. Recommendations for how this type of synthetic data could be incorporated into modeling practice will conclude the presentation.
This material is based upon work supported by the National Science Foundation under Grant No. 1415924. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.