By: Laura Schewel on January 18th, 2017

Print/Save as PDF

Does Transportation Data Actually Matter?

Big Data  |  Transportation

At last week’s Transportation Research Board Annual Meeting, I attended an excellent panel discussion about transportation data. Towards the end of the panel, the moderator challenged the group with a somewhat loaded question. I don’t recall the exact phrasing, but it was along these lines: “As transportation professionals, we know that we have a huge amount of work to do to upgrade, maintain, and repair our infrastructure. The backlog of projects that we have not yet begun is overwhelming. Given our clear mandate - and the often politicized process of infrastructure investments - does all this new data actually impact our decisions in the real world?” It’s an important question to ask - but based on audience feedback and in my own opinion, the answer is clearly yes.

Why Transportation Data Matters

For this blog post, I've collected a few of the ways that data does make a difference for transportation. As noted in the "Sources" section below, some of these responses are paraphrased from audience and panelist remarks, and some are based on my own experience providing data for hundreds of infrastructure and planning projects.

 

1. Rapidly Changing Communities

Transportation behavior is changing rapidly, and so are key factors such as household characteristics; for example, millennials living at home after college or buying fewer cars than their parents. We have to take that behavior and its rapid rate of change into account when allocating dollars to projects.

However, many of the decisions made today are based on household surveys that were taken when companies like Uber and Lyft didn’t exist. Perhaps projects that were obviously top priorities six years ago are no longer as important today - but we need current data to know with certainty that our priorities should change.**

 

2. Better Forecasts

According to Harvard Business Review, bad data costs the US economy $3.3 trillion every year - and the old adage of “bad data in, bad data out” applies to transportation, too. Better data will improve our forecast models by helping to ensure that our baseline understanding of current behavior accurately reflects reality.

We still rely on these models to prioritize between the many competing needs in our infrastructure, whether it’s repair, maintenance, or expansion projects. Misspecified models lead to misallocation of scarce resources.* When our data is inaccurate, our policy and infrastructure decisions are far more likely to perform poorly.

 

3. We Can Do Better

We want to do better in the future than we’ve done in the past! No one wants to sit in traffic, but a congested commute is the reality for many American workers today, especially those in major metropolitan areas. For example, as we make investments in our transportation infrastructure, new data resources can add more richness around the ways that transportation infrastructure impact different demographic groups. Thus, we can think more rigorously about equity when making decisions. This is a clear instance of how better data could change the political discourse around investment.*

Another audience member later pointed out - and rightly so - that rural areas are often under-measured and thus under-represented in transportation decision making. New Big Data resources offer a cost-effective way to get them equal measurement and stronger footing in the conversation.*

 

4. We Can Change Minds with Data

One more experienced audience member commented that he’d spent most of his career in “an era of numberless planning,” which I thought was an elegant description. He said in the new, data-rich era we’ve entered, empirical data can really can help change the mind of a policy maker or stakeholder - but only if it is presented well.*

Similarly, if an old or different data set is incorrect or biased, having multiple data lenses through which to look at a question can help hedge against incorrect or biased data. Measuring travel patterns empirically can help prevent policymakers and civic leaders from relying soley on the one source that justifies their "favorite" project.**

 

What Do You Think?

All said, if we accomplish nothing more in the next five years except for the backlog of construction and maintenance projects that we’ve already identified, and pay no real attention to prioritization (which is a bad idea!), we would still need current transportation data to help plan effective detours and construction-misery mitigation tactics.

I hope to add to these list in the future and hope to hear more ideas from our clients and partners. Please check in in coming months for more additions to this list, and leave a comment below to share your ideas!

 

Sources

*Signifies a response I heard from an audience member

**Signifies a response I heard from a panelist 

This blog post is a reaction to a panel discussion convened at TRB 2017. “Data: What Difference Does it Really Make?” The keynote presenter was Aly M. Tawfik from Cal State. The panelists were Steve Polzin from the University of Southern Florida, Erik Sabina from Colorado DOT, Stacey Bricka from MacroSys, and David Hartgen from the Hartgen Group. 

 

big data transportation