Area of Interest 05- Decision Analytics, Operational Planning and Predictive and

Probabilistic AI Modeling

ATTACHMENT 1: SOLICITATION TOPICS AREAS/AREAS OF INTEREST

1.0 Problem Statement

DLA has an urgent demand for improved visibility of the supply chains to respond proactively. DLA requires artificial intelligence (AI) capability to interact with critical operational data in a classified environment leveraging DLA's existing Gemini-AI based large language model (LLM).

2.0 Background

As the nation’s logistics combat support agency, the Defense Logistics Agency (DLA) manages the end-to-end global defense supply chain – from raw material to end user disposition – for the five military services, 11 combatant commands, other federal, state and local agencies, and partner and allied nations.

DLA’s mission is to “deliver readiness and lethality to the Warfighter Always and support our nation through quality, proactive global logistics.” To accomplish this mission, DLA has a staff of about 25,000 employees divided into multiple supply chains that contract for materiel and services across the military classes of supply: subsistence (food/water), clothing and textiles, bulk petroleum and other energy products, construction materiel and equipment, personal demand items, medical materiel and equipment, and repair parts for land, sea and air systems.

(Defense Logistics Agency, 2024)

DLA J6 provides program management, information technology, and technical expertise to enable secure, scalable digital capabilities across the enterprise. As operations become more complex and unpredictable, DLA requires enhanced capabilities to improve data visibility, integration, and decision making.

The current analytical tools are insufficient and lack the necessary interaction with data to enable decision making based on probabilistic and predictive models. Additionally, current processes and solutions do not provide timely access to logistics data sourced from enterprise resource planning (ERP) systems and other operational data repositories, causing delays in decision making, reducing operational efficiency. Current systems do not allow personnel to interact effectively with critical operational data in classified environments or leverage predictive and probabilistic insights. To address these gaps, DLA is pursuing the rapid deployment of a

Gemini-based prototype to provide the agency with improved access to and interaction with critical, operational data – including data derived from ERP systems and related logistics platforms, in both SIPR and NIPR environments. This capability leverages DLA’s existing

Gemini-AI LLM, which serves as the foundational user interface and reasoning engine, enabling personnel to interact with data using natural language queries to generate insights and support decision making. This capability empowers personnel to anticipate and mitigate risk, respond proactively to disruptions, and maintain continuous support even in challenging or contested operational environments. DLA needs AI-enabled solutions that improve access to and understanding operational data, supporting proactive decision making and mission readiness in a complex environment.

3.0 Desired Solution Features

3.1 Secure access to classified and operational data.

3.2 Compatibility and Integration with the existing Gemini-AI LLM.

3.3 Predictive and probabilistic analytics and modeling to anticipate issues.

3.4 Near real time access and processing of information.

3.5 Ability to unify data from dozens of systems into a clear picture

3.6 Transparent and explainable AI outputs to support decision-making.

3.7 Scalable design for use across the agency

4.0 Project Expectations

4.1 Deliver a working minimum viable prototype (MVP) that leverages DLA's existing

Gemini-based LLM capability in 2-4 weeks.

4.2 Demonstrate functionality and ability to quickly deploy a prototype in classified and controlled IL5 and IL6 environments. Include a high-level timeline that shows rapid development, testing, and deployment.

4.3 Integrate with representation data as permitted.

4.4 Collaborate with government staff to refine capabilities.

4.5 Show measurable improvements in data visibility, interaction with data and AI output to assist in decision making.

5.0 Intellectual Property

At a minimum, DLA anticipates acquiring Intellectual Property:

5.1 Software Rights

5.1.1 Any software delivered, including commercial licenses customarily provided to the public.

5.2 Technical data for the solution:

5.2.1 Data necessary for operation, maintenance, or training.

5.2.2 Data describes the required overall physical, functional, and performance characteristics (along with the qualification requirements, if applicable) of an item, component, or process to the extent necessary to permit identification of physically and functionally interchangeable items. This is often called ‘form, fit, and function’ data (see 48 C.F.R. § 252.227-7015(a)(3)).

6.0 Foreign Ownership and Control; Foreign Access to Technology

6.1 The Responder shall provide details of any foreign ownership or control that may exist within their corporate structure.

6.2 The Responder shall have the ability to create and store export-controlled data in accordance with export control regulations.

7.0 Rough Order of Magnitude (ROM)

Proposer solution brief must provide a ROM estimate for the project expectations in section 4 above.

Project Expectations Estimate

4.1 – $

4.2 – $

4.3 – $

4.4 – $

4.5 – $

TOTAL $

8.0 Submission Instructions

Solutions Briefs are due no later than April 10, 2026, at 5:00PM EDT. Please submit the

Solution Briefs via email to Thomas Walsh Thomas.Walsh@dla.mil.

ATTACHMENT 1: SOLICITATION TOPICS AREAS/AREAS OF INTEREST