This study had the following objectives:

1. To identify and clarify these two emerging effects – real time data and changing culture,

2. To identify the shifts in data collection and transportation modeling that must take place to

assist in identifying and forecasting travel behavior, and

3. To discuss the impacts of such operational shifts, both in cost and outcomes to provide NYMTC with the cost and efficacy impacts of incorporating these

emerging tools.”

To address these objectives, the research team at Albany Visualization and Informatics Lab (AVAIL), led by Dr. Catherine Lawson, PhD., from the University at

Albany, conducted a literature review; a cost benefit analysis of current and emerging transportation data surveying and modeling

methodologies; and produced a set of recommendations for the near-term and the longer-term.

The literature review and cost benefit analysis revealed certain facts about the state of travel data collection in the United States. The paper travel diary remains the

predominant instrument for collecting travel data despite its well-documented shortcomings and high cost. GPS devices have

grown in popularity but are used primarily as a supplement for paper travel diaries. The value of travel surveying via smartphones is no longer a strictly academic

question as numerous agencies have used the smartphone in a travel survey, either as the primary survey instrument or in a

subsample (Indiana, Oregon, Singapore, Boulder, etc.). Origin-destination tables used in travel demand models can be constructed from social media posts and calldetail

records though these datasets often lack valuable information (such as reliable trip purpose or mode), present

incomplete pictures of travel (especially in the case of social media), are prohibitively aggregated, or are not representative. Finally, the research team found it is

possible to develop and deploy a travel demand model that does not use travel survey data for inputs, opening the possibility of eliminating or reducing the size of

travel surveys.

This study introduces two categories to classify data collection efforts:

1. Active Data Collection – the use of self-report and surveying to generate data.

2. Passive Data Collection – the acquisition of existing data.

The research suggests three orientations toward travel data collection, each with their own risks and advantages that could satisfy NYMTC’s modeling needs, while

enabling future cost savings and/or increases in data quality. Two of these three orientations (or pathways) emphasize Active Data

Collection strategies while the third emphasizes Passive Data Collection. These pathways are fluid and dynamic. They are not intended as a step-by-step guide to the

future. Instead, they are intended to illuminate the data collection trajectory, highlighting opportunities and delineating the consequences, both positive and

negative, of various data collection decisions.

Briefly, these pathways are:

1. Paper and Online Diary Travel Survey (with GPS or Smartphone Supplement)

2. Smartphone Diary Travel Survey (with Online Supplement)

3. Passive Data Collection (with Smartphone Supplement)