Insight by Hitachi Vantara Federal

Take a programmatic approach to data analytics

Having acknowledged the importance of data in decision-making, and having hired chief data officers and data scientists, many agencies are at the “now what?” stage. Recent statutory and policy development mandates data-driven decision-making and the related activities comprising data management. Program managers, data staffs, and technology sections are working to operationalize data-related policies.

For a sense of what challenges they face and some promising practices for reaching their goals, we asked a trio of experts with an external yet government-wide view.

The primary challenges aren’t technical, Hix said. Rather, agencies face the bureaucratic hurdle of how long it takes to acquire the latest tools from industry. It causes the risk of buying outdated technology by the time agencies complete the procurement cycle. Hix said this is why many turn to the faster Other Transaction Authority procurement methodology for prototyping solutions to data requirement.

Another challenge is the human one.

According to Konnert, agencies often “lack having the advocates … that understand the value of the data, and can help you drive the value out of that data.”

Not that the technical challenges are simple. A practice to avoid, Nayak said, is basing a data analytics project on the tools available, rather than on the specific problem at hand. Agencies need “a shared language with which the technologists can talk to the stakeholders, to the decision makers, and to the people who are actually using the system.”

She cited a Navy example, where the problem to be solved was the proper interval for cleaning ships’ hulls. It’s an expensive and time-consuming task, yet it’s crucial to efficient operations because collected barnacles and other material seriously degrade ship performance.

By curating the right data sources and applying the right analytics tools, Nayak said, Navy base planners discovered that more frequent cleaning lowered operation costs because the increase in fuel efficiency outweighed the costs of scrubbing hulls.

Check out the video for more insight into designing and executing effective data analytics programs aimed at improving mission delivery.

Data Science Challenges

Analytics success “becomes a decision engineering problem; getting the data owners and the mission owners together so that they are using data science and analytics as a decision-making tool."

Data Analytics Use Cases

I think one of the things that we want to get to, and I think the federal space needs to get to, is collaboration and sharing of that data across agencies. And that's where that architecture that you put in place becomes even more important, so you have common standards between the different agencies.

Listen to the full show:

Featured speakers

  • Dr. Pragyansmita Nayak

    Chief Data Scientist, Hitachi Vantara Federal

  • Gary Hix

    Chief Technology Officer, Hitachi Vantara Federal

  • Kerry Konnert

    Senior Director, Services and Delivery, Hitachi Vantara Federal

  • Tom Temin

    Host, The Federal Drive, Federal News Network

Sign up for breaking news alerts