Logistics Prediction for Discontinued Military Supplies & Procurement

Overview

Uncertainties in the discontinuation and procurement volume of military supplies can lead to lower equipment readiness, increased costs, and threats to national security. Currently, logistics data managed individually by the Army Logistics Command, the Ministry of National Defense (MND), and the Defense Acquisition Program Administration (DAPA) is fragmented, making integrated analysis difficult. This project aims to build a system that centrally manages military logistics data and utilizes AI-based predictive models to preemptively address procurement delays and item obsolescence.

Project description

In this project, various public and private datasets were collected and integrated to develop AI predictive models. Specifically, we standardized procurement and discontinuation data sourced from public agencies (Logistics Command, MND, DAPA) and defense contractors to create a reliable environment for data-driven forecasting. Machine learning and data analysis techniques were then applied to optimize models for predicting discontinued parts and required procurement volumes.

Considering the sensitive nature and security requirements of military data, we established a Federated Data Lake system that can be operated internally within the Logistics Command for data processing and training. By utilizing the DPG AI/Data Lake for performance evaluation, we derived the optimal predictive algorithms. The resulting "Discontinued Item & Procurement Quantity Prediction Service" is provided via API, with restricted access granted to defense contractors to support their operational planning.

Key technological features

• ETL/ELT Process Design: Automation of data Extraction, Transformation, and Loading (ETL/ELT) processes
• Data Standardization: Ensuring consistency and compatibility across disparate data sources for seamless integration
• AI Model Development & Optimization: Development of predictive models for discontinued items and procurement quantities, followed by optimization through hyperparameter tuning and rigorous model validation.

Expected outcomes

By preemptively forecasting the discontinuation of military supplies and the necessary procurement quantities, this system helps maintain high equipment readiness and prevents unnecessary cost increases. Previously, the lack of systematic management for discontinued parts led to procurement delays and operational disruptions; however, this system now enables proactive responses.

Furthermore, by providing this critical information to defense contractors, the project facilitates more efficient production and procurement planning, which is expected to revitalize the defense industry. Ultimately, this AI-driven approach enhances the stability of the military supply chain and strengthens national defense capabilities through data-based logistics management.

Launch year: 2024

Attribution

Case study courtesy of the National Information Society Agency, Republic of Korea, Apolitical's Content Partner. Sourced from Use Cases of Public AI Service, Vol.2, https://eng.nia.or.kr/site/nia_eng/ex/bbs/View.do?cbIdx=31975&bcIdx=29182&parentSeq=29228