AI 기반 전시 군수소요 예측을 활용한 선제적 지속지원체계 발전방향

김형곤^*

해군 소령/해군대학 지휘참모과정 학생장교

Development of an AI-Enabled Proactive Sustainment System Using Wartime Logistics Demand Forecasting

Hyeong Gon Kim^*

LCDR, ROK Navy/Student officer, Naval Commander & Staff Course, ROK Naval War College

Correspondence to: *Hyeong Gon Kim 271, Jaun-ro, Yuseong-gu, Daejeon, Republic of Korea E-mail: jandane@naver.com

초록

현대전에서는 전투의 속도와 전력의 분산 정도가 증가하면서 적시적인 지속지원이 더욱 중요해지고 있다. 특히 전선이 북상할 경우 열악한 수송환경과 길어진 병참선으로 인해 후방의 군수품이 전방부대에 도달하기까지 상당한 시간이 필요할 수 있다. 현행 전시 군수지원은 사전에 산정된 전시소요를 기반으로 준비하면서도, 실제 전투에서는 전방부대의 소요제기에 따라 필요한 물자를 보급하는 방식이 병행된다. 이 과정에서 소요판단, 분류, 포장 및 수송에 시간이 소요될 경우 작전템포와 전투력 유지에 영향을 줄 수 있다. 본 연구는 이러한 제한을 보완하기 위해 AI를 활용하여 부대별·작전단계별 군수소요를 사전에 예측하고, 전투 중 실제 소모정보를 반영하여 예측값을 지속적으로 보정하는 선제적 지속지원 개념을 제안한다. 또한 예측된 소요를 전방 분배허브의 물자 준비와 부대별 군수지원 패키지에 연계함으로써 소요제기 이전에 필요한 군수품을 준비하고 적시에 지원하기 위한 발전방향을 제시한다.

Abstract

Modern warfare requires increasingly timely sustainment as operational tempo accelerates and forces become more dispersed. When ground forces advance, deteriorating transportation infrastructure and extended lines of communication may significantly increase the time required to deliver supplies to frontline units. This study proposes an AI-enabled proactive sustainment concept that forecasts logistics demand by unit and operational phase in advance, and continuously updates the forecast using actual consumption data gathered during operations. The forecasts are linked to forward distribution nodes and pre-configured sustainment packages so that critical supplies can be prepared in advance before formal requests are submitted. The proposed concept is intended to complement, rather than replace, the existing request-based logistics system and to reduce sustainment delays that may affect operational tempo and combat endurance.

Keywords:

Artificial Intelligence, Sustainment, Wartime Logistics, Predictive Logistics, Proactive Sustainment

키워드:

인공지능, 지속지원, 전시 군수소요, 예측군수, 선제적 군수지원

References

Seunguk Kang, Gakgyu Kim, Seohee Jang, & Jaehyung Bae, ‘The Study on the Improvement of Wartime Resource Requirements Analysis,’ in proceedings of 2024 Korea Institute of Military Science and Technology Annual Conference, 2024, pp. 248-249.
Byeonghee Jeon, Jeonghyeok Lim, & Seongam Moon, ‘Quantitative Calculation of Optimal Supply Level for Types of Wartime Demand: Focusing on Army Fuel Supply,’ Logistics Research, VOL. 32, NO. 6, 2024, pp. 1-16.
Sangbaek Kim, Bora Kim, & Honggu Moon, ‘Improvement of Wartime Requirement Estimation System Reflecting Winter Operational Environment: Focusing on Spare Parts, Ammunition, Combat Rations, and Medical Supplies,’ KIDA Defense Issues & Analyses, NO. 2091, 2026.
Byung-Ho Jung & Ik-Ki Kim, ‘Methodology of Location Decision for a Transfer Center Distributing a Necessary Resource Item while Considering Characteristics of the Material in Wartime,’ Journal of the Korea Defense Management Analysis Association, VOL. 35, NO. 2, 2009, pp. 1-14.
Min Suk Kim, Whan Sik Jung, & Jae Yeong Lee, ‘A Study on VV&A Application for the Korean Wartime Resource Requirement Model,’ Journal of the Korea Defense Management Analysis Association, VOL. 35, NO. 2, 2009, pp.48-58.
Kyuyong Shin, Jong-Kwan Lee, Kwang-Hee Kang, Wongi Hong, & Changhee Han, ‘The Current Applications and Future Directions of Artificial Intelligence for Military Logistics,’ Journal of Digital Contents Society, VOL. 20, NO. 12, 2019, pp. 2433-2444. [https://doi.org/10.9728/dcs.2019.20.12.2433]
Jaehun Jung & Taesoo Sim, ‘A Study on Improving the Accuracy of Air Force Spare Parts Demand Forecasting Using Machine Learning and Deep Learning,’ Journal of Product Research, VOL. 43, NO. 3, 2025, pp. 67-73.
Yong Kim, Ji-Hun Kang, & Kyeong-mo Son, ‘A Study on the Promotion Directions of Data-Driven Logistics Innovation,’ The Journal of the Convergence on Culture Technology, VOL. 11, NO. 3, 2025, pp. 199-206.
Euna Lee & Yongmin Kim, ‘Introduction of Explainable Artificial Intelligence in Defense Logistics,’ KIDA Defense Issues & Analyses, NO. 1892, 2022.