Knowing what’s happening in our oceans and waterways has critical implications globally. Whether it’s monitoring international trade and shipping, determining environmental hazards, or tracking illegal activities, understanding maritime environments is essential.
Together, this is known as Maritime Domain Awareness (MDA). MDA is the effective understanding of anything associated with the global maritime domain that could impact the security, safety, economy, or environment of the United States.
This opportunity to achieve awareness on a global scale prompted NS2 Labs to start a new maritime-focused project. Here’s a quick summary of the trends, technology, and data NS2 Labs explored this past year as we kicked-off our inaugural MDA project.
Diving into Maritime Domain
NS2 Labs evaluates new projects based on whether there is a good degree of alignment between what the problem requires and the types of solutions we can help provide. In the case of delving into the world of MDA, the project hit all the right points:
- Overt interest from the Navy, the DHS, two IC customers, and a NATO partner
- Support from industry partners who have broad domain expertise and access to the right data
- Existing technology solution—SAP HANA—seemed well situated to solve the problem
- Corporate interest from SAP itself (with 70% of the world’s supply chain data on SAP systems there is extensive corporate expertise in this space)
After evaluating customer and partners’ relevant use cases, Dark Vessel Detection and Surface-to-Surface Transfers were selected for initial implementation. NS2 Labs then started building out a technical solution for each of the use cases.
SAP HANA for MDA
Our running hypothesis was that the SAP HANA platform would be well suited for MDA analytics:
- As an in-memory data analysis platform, utilizing the platform would allow high-speed analysis.
- The platform contains integrated engines that allows for rapid deployment of capabilities and a compact form factor for disconnected operations.
- It also supports data streaming across these engines, so whatever analytics were developed could immediately be applied to real-time data as well.
To test this approach, NS2 Labs obtained a mixture of AIS and radar data from customers and partners, and loaded it into the HANA framework.
Native spatial and machine-learning engines were used to apply LDA, DBSCAN, and VRNN algorithms to the data and were quickly able to address the identified use cases. We were able to detect dark vessels with a high degree of reliability and identify surface transfers while minimizing the identification of benign surface contacts.
While this proved the technical approach was viable, we knew this wasn’t enough.
The Next Phase
Dark Vessel Detection and Surface-to-Surface Transfers algorithms exist on several commercial programs already, so we needed to develop an approach that would offer something new moving forward. And that’s where private and public sector partners come in.
The next phase for this MDA project promises to be the most exciting because of the groundwork already in place:
- Platform with compelling data and established relationships with “non-traditional” data providers
- High fidelity port information, including physical, logistical, and staff modeling
- Data science team who has already defined the algorithms to identify “normal” shipping behaviors, thereby allowing predictions of future behaviors
- Additional relevant use cases
- Full funding from NS2
What NS2 Labs needs now is YOU. If you are part of a government or public organization that has interest and the expertise in this type of project, we would love to hear from you.
Please contact us at https://www.sapns2.com/labs/ and help us leverage new and emerging technologies to improve the future of Maritime Domain Awareness.
