Accelerating Innovation with AI Using Synthetic Data

Proprietary systems. Manual data entry. Data sensitives and volume. Legacy mindsets.

With the use of synthetic data, Vibronyx has overcome such hurdles, accelerating software development and building AI models for a Defense Logistics Agency (DLA) supply chain Analytics Command Center pilot solution.

Vibronyx CEO Clayton Nicholas recently shared insights and learnings from this customer success story at the annual Nashville Analytics Summit.

What is synthetic data?

Vibronyx Chief Technology Officer Vishnu Venkatesh calls synthetic data “the underutilized workhorse in the data community. We use it to overcome barriers with using real data for AI projects to be more efficient and cost effective. Our work with the DLA has allowed us to pioneer a platform that improves supply chain resiliency and agility in an ecosystem with complex data and data security requirements.”

Gartner predicts, by 2030, synthetic data will “overshadow” real data in AI models. In its white paper on how synthetic data will be the future of AI, Gartner analysts Leinar Ramos and Jitendra Subramanyam write “…the fact is you won’t be able to build high-quality, high-value AI models without synthetic data. Synthetic data is on a trajectory to go from a sideshow to becoming the main force behind the future of AI.”

According to the Harvard Business Review, the business value of leveraging synthetic data includes accelerating product development and testing, enables faster data sharing and improves efficiencies.

The business benefit, Nicholas says, is a lower cost, more efficient solution.

“Synthetic data can speed up product and analytics development and testing cycles, lower the cost of data acquisition, bridge information silos, support regulatory compliance, preserve privacy and improve data governance.”

Data can be generated using different methods from simple to sophisticated based on the use cases and how closely the synthetic data needs to represent real data.

use-case complexity spectrum

Synthetic data and the DLA

Vibronyx’s partnership with the DLA is a multi-phase project. The team quickly realized complex strategic, political and geographical hurdles were going to impact project timelines and data sharing. There were also external factors – like increased cyber-attacks – requiring a greater vigilance and scrutiny.

The DLA manages one of the largest Global Supply Chains across nine classes of supply, with daily averages of 89K orders. This equates to: $41B per year in sales and 27K employees.

Future State vision resulted in a set of agreed-upon and prioritized innovation use cases for the DLA and Military Services.

synthetic data innovation use cases for the DLA and Military Services

Synthetic data allowed the Vibronyx team to develop:

  1. Prototyping for software development;
  1. AI training models; and
  1. Building domain knowledge and analytics use cases from integrated supply chain data.

Generated synthetic data improved some of the greatest hurdles the team faced: data quality, data security and privacy, improve data distribution, and data sharing.

“We used integrated synthetic data across the supply chain to engage stakeholders and SMEs to get an end-to-end; data-driven supply chain view for the first time – breaking down silos,” said Nicholas. “We went from nearly ending the pilot program to building something that bridged gaps in a historically challenging, politically charged bureaucracy. We’re proud of our work to date and look forward to the next phase of our work with the DLA.

The result?

Improved Supply Chain Resilience, Agility, Warfighter Readiness and Customer Satisfaction.


Contact Us

Questions about our digital supply chain solutions? Let’s talk.

Read the Latest

If you enjoyed reading this, then please explore our other articles below:

Go to Top