Page 28 - FCW, October 2017

DATA ANALYTICS
Executive Viewpoint
DATA DRIVES DOT MISSION
Harnessing massive volumes of disparate data is a significant but valuable challenge.
Nearly every government agency is awash in huge volumes of increasingly varied data. Everyone acknowledges that data can help any agency fulfill its mission, but deriving value from an increasingly disparate and dynamic dataset is a challenge. GCN caught up with Daniel Morgan, Chief Data Officer at the Department of Transportation, to hear his thoughts on data analytics and how the DOT can best harness data to drive its mission forward.
What does data mean for the DOT?
We have a varied mission at the DOT. We invest in, monitor, and regulate the national transportation system: we use data to manage the safety and operations of the national airspace; and we also operate the St.
So when I think about predictive analytics, sometimes it’s a matter of measuring risk, sometimes it’s a matter of understanding some challenges with regulated entities. Sometimes there are even fraud schemes. For instance, we need to estimate the risk that a new registrant is not just a reincarnation of a previous registrant.
When it comes to data analytics, are you trying to get more information from the data you already have, or to get a better idea of other data you need?
There are a couple of things happening at the DOT. Do we
want to make better use of the data we have? Absolutely. Transportation happens within both an economic and a social
SPONSORED CONTENT
Lawrence Seaway with our partners in Canada.
“It turns out we need more and different data to measure those kinds of risk factors, and not all of it lives inside the DOT.”
Unlike the corporate
sector, our data is truly
varied. Some of it is tabular;
but it also includes images,
video, and spatial data. That
makes for an interesting
data-management challenge, but we have to use data every day in our decision-making. It informs how we approach delivering our programs to ensure we have a safe and efficient transportation system that serves everybody.
context. It’s based on how we use the environment, how we zone housing, and how all of these and other land use characteristics drive how people interact with the environment. So I have
to bring in other datasets to build context around the things
I’m observing in my own transportation data. I have to start building correlations worth exploring to help us understand the extent to which our programs are working and where we might better invest our time.
Has the new generation of analytical tools given you any idea of new applications and services you can use to further the DOT’s mission?
It’s not really about new tools. There are decades-old ideas about what drives data analysis that are still applicable today. One
is the increasing desire for quantification in a wider range of fields. Turning text into numbers means you can use that for data analytics. Similarly, with sound and video, there are a bunch of things you can do when you turn that data into numbers.
Then there is the ever increasing power of computing. We no longer need a supercomputer or mainframe to do some of
What do you think about such things as predictive analytics and how that can help?
It depends on what you’re trying to predict. One thing I’m interested in is measuring risk in the transportation system.
The number of roadway fatalities over the last two years went up eight percent – each year. We went from losing around 32,000 lives just a couple of years ago, to 37,000 lives in 2016.
That census by itself isn’t good enough for us to understand whether risk is changing in the environment. Are people driving more? Why are they driving more? Are their jobs in different places? Are there new vulnerabilities in the system? Or have travel patterns changed to the extent that we’re seeing old vulnerabilities just now being revealed? It turns out we need more and different data to measure those kinds of risk factors, and not all of it resides inside the DOT.