You are here

Research and Testing of Artificial Intelligence (AI) at Defense Logistics Agency (DLA) Distribution Center Warehouses


RT&L FOCUS AREA(S): Artificial Intelligence/ Machine Learning

TECHNOLOGY AREA(S): Information Systems

The technology within this topic is restricted under the International Traffic in Arms Regulation (ITAR), 22 CFR Parts 120-130, which controls the export and import of defense-related material and services, including export of sensitive technical data, or the Export Administration Regulation (EAR), 15 CFR Parts 730-774, which controls dual use items.  Offerors must disclose any proposed use of foreign nationals (FNs), their country (ies) of origin, the type of visa or work permit possessed, and the statement of work (SOW) tasks intended for accomplishment by the FN(s) in accordance with section 3.5 of the Announcement.  Offerors are advised foreign nationals proposed to perform on this topic may be restricted due to the technical data under US Export Control Laws.

OBJECTIVE: Develop an innovative Artificial Intelligence (AI) solution with a state-of-the-art capability that operates within the DLA Distribution Warehouse environment. The warehouse AI system may use various sensors (e.g., Internet of Things (IoT)) where applicable. It should minimize the need for infrastructure modifications to enable an artificial intelligence system within the warehouse environment. The goal of this objective is for the vendor to develop a capability for a warehouse AI system that addresses the requirements for integration with a Warehouse Management System (WMS) and a Warehouse Execution System (WES) as specific warehouse infrastructures dictate. This capability will provide for the seamless execution of AI and interactions with Smart Warehouse systems such as 5G Networks, IoT Sensors, Blockchain technology, Quantum Computers, and Machine Learning (ML).

The state-of-the-art AI solution must integrate into the existing warehouse communications systems to communicate with WES systems when installed. This integration allows Autonomous Guided Vehicles, Autonomous Mobile Robots (AMRs), Robotic Arms, IoT Sensors to receive tasking in an automated fashion to operate frequently and report success or failure at tasking. In support of routine warehouse operations, this research seeks to identify and test AI technology that can be used uninterruptedly and continuously within the DLA Distribution Warehouse environment. This research effort addresses DLA identified cybersecurity requirements through the test and evaluation of government security controls. It leverages current technologies in the AI industry. This research project will operate in locations at designated DLA Distribution Centers in the United States.

DESCRIPTION: Defense Logistics Agency (DLA) Distribution Modernization Program (DMP) topics of interest are research focused on a Continental United States-based Artificial Intelligence (AI) solution in support of the routine warehouse operations. This research project will involve the use of Commercial/Industry AI technology that can meet the demands of warehouse operations, can be integrated with autonomous warehouse vehicles, robots, and warehouse communications, and be integrated with warehouse navigation systems, 5G Networks, IoT Sensors, Quantum Computing architecture, and warehouse based Machine Learning (ML) that:

1. Support a joint effort between DLA Research and Development (R&D) and DLA J4 Distribution Headquarters to conduct research and test a warehouse AI system that works with various autonomous platforms, 5G Networks, IoT Sensors, Quantum Computing systems, and ML applications during warehouse operations.

2. Significantly addresses the AI capabilities of AI within a distribution warehouse operations environment.

3. Features an AI system able to implement high precision data for regular use in warehouse operations.

4. Can be integrated into warehouse communications systems such as a WMS or a WES to receive tasking and report status.

5. Demonstrates a state-of-the-art operational capability when operating within the distribution warehouse environment through the application of AI technology and facilitates a robust communications network technology used in a working environment shared with warehouse workers.

6. It is a reliable and robust technology solution that allows DLA Distribution Warehouses to perform automated tasks without significantly lower operating speeds per existing industry trends.

7. Demonstrates compatibility with a Government data cloud environment to store and retrieve warehouse-generated data without relying on a separate commercial data cloud environment to navigate successfully.

8. Conclusively demonstrates the use of new AI technology and concepts for application and integration in the distribution and delivery of material and goods during representative distribution warehouse operations in an innovative way.


PHASE I: NTE 12 Months $150K- Base NTE $100K base 6-9 Months, - Option 1 NTE $50K base 3-6 Months

PHASE II: – NTE 24 Months $1.6M - Base 12-18 months, $1M Option 6 Months NTE $.6M

PERIOD OF PERFORMANCE: The phase one period of performance is not to exceed 12 months total.  Options are not automatic.  Approval is at the discretion of the DLA SBIP Program Manager.  The decision is based on Project Performance, Priorities of the Agency, and/or the availability of funding.

PHASE I: Perform a design study to determine how to use artificial intelligence to optimize DLA Distribution Warehouse operations, sustainment, and logistics support. Deliver a final design of AI's capabilities, a simulation model of DLA Distribution assets, and a demonstration of an AI-infused model capable of making intelligent trade-off decisions to meet specified PM requirements. A successful design will optimize support, minimize DLA Distribution Warehouse system downtime, and maximize system availability, using logistics inputs (component failure rates, shipping times, repair times, maintenance man-hours, and warehouse staffing).

The research and development goals of Phase I provide Small Business eligible Research and Development firms the opportunity to successfully demonstrate how their proposed warehouse AI concept of operations (CONOPS) improves the distribution of goods and materials within the DLA distribution enterprise and effectively lessens the time to provide needed supplies to the Warfighter. The selected vendor will conduct a feasibility study to:

1. Address the requirements described above in the Description Section for warehouse AI operations.

2. Identify capability gap(s) and the requirement for DLA to use AI in the DLA Distribution Operations environment.

3. Develop the vendor's Concept of Operations (CONOPS) to utilize warehouse AI and describe clearly how the requirements develop from it.

Note: During Phase I of the SBIR, testing is not required.

The vendor must create a CONOPS for Warehouse AI in support of both routine and wartime distribution warehouse operations. The concept of operations will cover the utilization of artificial intelligence within distribution warehouses during routine procedures, describing precisely all operational requirements as part of this process. This artificial intelligence requirement intends to operate inside distribution warehouses successfully.

The deliverables for this project include a final report, including a cost breakdown of courses of action.

PHASE II: Based on the research and the concept of operations developed during Phase I, the research and development goals of Phase II emphasizes the execution of the Warehouse AI system following the typical DLA Distribution Warehouse concept of operations for materiel handling. During Phase II, the vendor will:

1. Address the specific user requirements, functional requirements, and system requirements as defined and provided by DLA.

2. Develop a prototype Warehouse AI system for Developmental Test and Evaluation (DT&E) and Operational Test and Evaluation (OT&E).

3. Implement government cybersecurity controls in the prototype design and secure all necessary cybersecurity certifications to operate the equipment in the DLA warehouse environment with DOD cloud connections.

4. Design the prototype equal to the technology maturity of Technology Readiness Level (TRL) 9 after Phase II.

5. Deliver a final Distribution Warehouse AI prototype system to DLA capable of successfully executing the operational concepts established in the Phase I CONOPS.

The DLA Warehouse Artificial Intelligence system will operate across the United States at various DLA Distribution Center sites mutually agreed upon between DLA R&D and DLA Distribution HQ. This project's deliverables include a final report, including a cost breakdown of courses of action (COAs).

PHASE III DUAL USE APPLICATIONS: PHASE III: Dual Use Applications: At this point, there is no specific funding associated with Phase III. During Phase I and Phase II, the progress made should result in a vendor's qualification as an approved source for a Warehouse Artificial Intelligence system and support participation in future procurements.

COMMERCIALIZATION:  The manufacturer will pursue the commercialization of the Warehouse Artificial Intelligence (AI) technologies and designs developed to apply to the warehouse environment. The processes developed in preliminary phases and potential commercial sales of manufactured mechanical parts or other items. The first path for commercial use will be at DLA's twenty-six Distribution Centers and twenty Disposition Centers. When fielded, DLA estimates 20 - 26 units, but the number of units could be more.


  1. Alex Krizhevsky and Geoffrey Hinton. Learning multiple layers of features from tiny images. 2009.
  2. Andrew G Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, and Hartwig Adam. Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861, 2017.
  3. Chunyuan Li, Heerad Farkhoor, Rosanne Liu, and Jason Yosinski. Measuring the intrinsic dimension of objective landscapes. Proceedings of ICLR, 2018.
  4. Devansh Arpit, Stanisław Jastrz˛ebski, Nicolas Ballas, David Krueger, Emmanuel Bengio, Maxinder S Kanwal, Tegan Maharaj, Asja Fischer, Aaron Courville, Yoshua Bengio, et al. A closer look at memorization in deep networks. In International Conference on Machine Learning, pp. 233–242, 2017.
  5. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770–778, 2016.
  6. Min, Hokey. (2010). International Journal of Logistics 13(1):13-39, “Artificial intelligence in supply chain management: Theory and applications.” February 2010. DOI: 10.1080/13675560902736537
  7. Misha Denil, Babak Shakibi, Laurent Dinh, Nando De Freitas, et al. Predicting parameters in deep learning. In Advances in neural information processing systems, pp. 2148–2156, 2013.
  8. Rosienkiewicz, Maria. (2013). "Artificial Intelligence Methods in Spare Parts Demand Forecasting. Logistics and Transport." 2013.
  9. Song Han, Jeff Pool, John Tran, and William Dally. Learning both weights and connections for efficient neural network. In Advances in neural information processing systems, pp. 1135–1143, 2015.
  10. Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324, 1998.
US Flag An Official Website of the United States Government