Lee, Jehun
Industrial and Systems Engineer
Profile
Lee, Jehun
Ph.D. candidate (KAIST)
Office: KAIST E2-1, 291, Daehak-ro, Yuseong-gu, Daejeon, Republic of Korea (34141)
Email: jehun.lee@kaist.ac.kr, swi02050@gmail.com
Research Interest
Operation optimization for smart factories
Autonomous decision systems
Scheduling for general systems
Reinforcement and imitation learning for scheduling
Education
Ph.D. in Industrial & Systems Engineering, KAIST, Korea, 2021.03.~
M.S. in Industrial Engineering, SKKU, Korea, 2016.09.~2021.02.
B.S. in Systems Management Engineering, SKKU, Korea, 2013.03.~2016.08.
Sejong Science High School, Korea, 2011.03.~2013.02.
Academic Works
2024
J.-H. Lee and H.-J. Kim, "Graph-based imitation learning for real-time job shop dispatcher". Under review.
2023
J.-H. Lee and H.-J. Kim, "Active schedule-based imitation learning for job shop scheduling", the Spring Conference of Korean Institute of Industrial Engineers, Jeju, Korea, 2023.
2022
J.-H. Lee and H.-J. Kim*, "Imitation learning for real-time job shop scheduling using graph-based representation," Winter Simulation Conference, Singapore, 2022, pp. 3285-3296. [Paper]
J.-H. Lee and H.-J. Kim, "Job shop scheduling using graph-based imitation learning", INFORMS Annual Meeting, Indiana, USA, 2022. [Poster]
H. Yang, J.-H. Lee, S. H. Lee, S. G. Lee, H. R. Kim, and H.-J. Kim*, "A multi-manned assembly line worker assignment and balancing problem with positional constraints," IEEE Robotics and Automation Letters, vol. 7, no. 3, pp. 7786-7793, 2022. [Paper]
J.-H. Lee, D.-Y. Kim, S.-H. Cho, and H.-J. Kim, "Reinforcement learning for resource leveling in multiple projects", the Spring Conference of Korean Institute of Industrial Engineers, Jeju, Korea, 2022.
J.-H. Lee and H.-J. Kim, "Dynamic job shop scheduling using graph-based imitation learning", the Spring Conference of Korean Institute of Industrial Engineers, Jeju, Korea, 2022.
2021
J.-H. Lee, H.-J. Kim*, Y. Kim, Y. B. Kim, B.-H. Kim, and G.-H. Chung, "Machine learning-based periodic setup changes for semiconductor manufacturing machines," Winter Simulation Conference, Online, 2021, pp.1-10. [Paper]
D.-Y. Kim, J.-H.Lee, S.-H. Min, S.-H. Cho, J.-H. Park, H.-S. Min, G.-W. Kim, M.-S. Kim, and H.-J.Kim, "Resource leveling in shipyard cargo hold process through reinforcement learning", the Autumn Conference of Korean Institute of Industrial Engineers, Seoul, Korea, 2021.
H. Yang, J.-H. Lee, and H.-J. Kim*, "Assembly line worker assignment and balancing problem with positional constraints," Advances in Production Management Systems (APMS), Online, 2021, pp. 3-11. [Paper]
H. Yang, J.-H. Lee, and H.-J. Kim, "Operation and optimization of the automotive parts assembly line considering worker skill levels," the Summer Conference of Korea CDE, Online, 2021.
2020
J.-H. Lee, Y. Kim, Y. B. Kim, B.-H. Kim, G.-H. Jung, and H.-J. Kim*, "A sequential search method of dispatching rules for scheduling of LCD manufacturing systems," IEEE Transactions on Semiconductor Manufacturing, vol. 33, no. 4, pp. 496-503, 2020. [Paper]
J.-H. Lee, Y. Kim, Y. B. Kim, B.-H. Kim, G.-H. Chung, and H.-J. Kim*, "A simulation-based sequential search method for multi-objective scheduling problems of manufacturing systems," Winter Simulation Conference, Online, 2020, pp. 1943-1953. [Paper]
H. Yang, J.-H. Lee, and H.-J. Kim, "Workforce assignment for automotive parts assembly lines," the Winter Conference of Korea CDE, Online, 2020.
K. T. Park, J. Lee, H. J. Kim, and S. D. Noh*, "Digital twin-based cyber physical production system architectural framework for personalized production," The International Journal of Advanced Manufacturing Technology, vol. 106, no. 5, 2020, pp. 1787-1810. [Paper]
H. Yang, J.-H. Lee, and H.-J. Kim*, "Workforce assignment with a different skill level for automotive parts assembly lines," Advances in Production Management Systems (APMS), Novi Sad, Serbia, 2020, pp. 206-212. [Paper]
2019
J.-H. Lee, Y. Kim, Y.-B. Kim, H.-J. Kim*, B.-H. Kim, and G.-H. Chung, "A sequential search framework for selecting weights of dispatching rules in manufacturing systems," Winter Simulation Conference, Maryland, USA, 2019, pp. 2201-2211. [Paper]
J.-H. Lee and H.-J. Kim, "A genetic algorithm for hybrid flow shop scheduling with multiple assembly stages," the Autumn Conference of Korean Institute of Industrial Engineers, Seoul, 2019.
2018
J. Kim, J.-H. Lee, H.-J. Kim, and B. D. Chung*, "Vulnerability analysis of evacuation transportation networks," International Journal of Industrial Engineering-Theory Applications and Practice, vol. 25, no. 5, 2018, pp. 663-673. [Paper]
J.-H. Lee, Y. Kim, J. Kim, Y.-B. Kim, H.-J. Kim*, B.-H. Kim, and G.-H. Chung, "A framework for performance analysis of dispatching rules in manufacturing systems," Winter Simulation Conference, Gothenburg, Sweden, 2018, pp. 3550-3560. [Paper]
J.-H. Lee, S.-S. Park, and H.-J. Kim, "Rescheduling algorithms for 3D printer-based manufacturing systems," the Summer Conference of Korea CDE, Jeju, 2018.
S.-S. Park, J.H. Lee, and H.-J. Kim, "Scheduling algorithms for 3D printer-based manufacturing systems," the Spring Conference of Korean Institute of Industrial Engineers, Gyeongju, Korea, 2018.
2017
J. Kim, J.-H. Lee, S.-L. Choi, H.-J. Jung, Y.-B. Kim, H.-J*. Kim, B.-H. Kim, and G.-H. Chung, "Rescheduling of flexible flow shop with sequence-dependent setup times and job splitting," Winter Simulation Conference, Las Vegas, USA, 2017, pp. 3745-3755. [Paper]
S.-S. Park, S.Y. Kim, J.H. Lee, and H.-J. Kim, "3D printer based assembly process scheduling algorithm development," the Winter Conference of Korea CDE, Pyeongchang, Korea, 2017, pp. 515-523.
Projects
Reinforcement learning for unrelated parallel machine scheduling problems with sequence-dependent setup times and machine eligibility
강화학습 기반 unrelated parallel machine 스케줄링 알고리즘 개발
셋업 및 설비 적합성 제약이 존재하는 병렬 설비 스케줄링을 위한 GNN 기반의 강화학습 알고리즘 개발
by Python
with KAIST, VMS Solutions
2023.07.~2024.06.
Production planning with AI
생산계획 최적화를 위한 AI 기반 방법론 개발
냉장고 생산 라인을 위한 생산계획 최적 수립 알고리즘 개발
by Python
with KAIST, LG Electronics
2023.07.~2024.02.
Graph-based reinforcement learning algorithm for real-time job shop scheduling
실시간 job shop 스케줄링을 위한 그래프 기반 강화학습 알고리즘
강화학습을 활용한 실시간 디스패칭 agent 개발
by Python (torch)
with KAIST (NRF of Korea)
2022.05.~2023.02.
Reinforcement learning for job shop scheduling
강화학습 기반 job shop 스케줄링 알고리즘 개발
강화학습을 활용한 실시간 디스패칭 agent 개발
by Python (torch)
with KAIST, VMS Solutions
2022.04.~2023.03.
Development of a wiring optimization algorithm for X-DEC slim layout
X-DEC slim layout 배선 최적화 알고리즘 개발
기판의 전극 위치가 주어진 상황에서 최적의 배선 도출 알고리즘 개발
by Python
with KAIST, SK Hynix
2022.03.~2022.03. (entire 2022.03.~2022.09.)
Reinforcement learning-based meta-scheduling for manufacturing systems
제조시스템 최적화를 위한 강화학습 기반의 메타 스케줄링 방법론 개발
다양한 스케줄링 문제에 적용할 수 있는 범용적인 agent 구축을 위한 meta-RL 적용
by Python (torch)
with KAIST (NRF of Korea)
2022.03.~ (entire 2022.03.~2025.02.)
Reinforcement learning for project scheduling
강화학습 기반 프로젝트 스케줄링
기존 작업 스케줄이 주어진 경우 작업자 부하의 분산을 최소화하기 위한 강화학습 기반 리스케줄링 알고리즘 개발
by Python (torch)
with KAIST, Samsung Heavy Industries
2022.03.~ 2023.07. (entire 2022.03.~2024.02.)
Development of a reinforcement learning algorithm for workload balancing of ship cargo production
강화학습을 통한 화물창 부하 최적화
기존 작업 스케줄이 주어진 경우 작업자 부하의 분산을 최소화하기 위한 강화학습 기반 리스케줄링 알고리즘 개발
by Python (torch)
with KAIST, Samsung Heavy Industries
2021.07.~2021.11.
Optimal machine assignment with machine learning algorithms
머신러닝 기반 최적 설비 대수 할당 모델 구축
반도체 FAB 생산라인의 생산성 최대화 및 셋업 시간 최소화를 위한 머신러닝 기반 주기적 할당 알고리즘 개발
by C# (MozArt), Python
with KAIST, VMS Solutions
2020.06.~2021.02.
Cyber-physical assembly and logistics systems in global supply chains
글로벌 공급 사슬의 조립 및 물류를 위한 사이버물리시스템
반자동 조립공정의 작업자 결근/이탈 상황에 대응 가능한 자동 대응 알고리즘 개발 및 빅데이터 분석 기반 설비 고장 예측
by Python
with KAIST, SKKU, Dexta, Yura, KTH, Scania, Ericsson, H&D wireless (EUREKA program, MOTIE of Korea)
2019.06.~2022.05.
Optimal weight sets for dispatching rules with multiple KPIs
다중 KPI를 고려한 디스패칭 룰 가중치 탐색 방법론 개발
스케줄 성능을 판별하는 지표가 다중인 경우 반도체 FAB 생산라인의 디스패칭 룰 기반 자동화 시스템의 최적 파라미터를 도출하기 위한 머신러닝 기반 알고리즘 개발
by C# (MozArt), R
with KAIST, VMS Solutions
2019.09.~2020.02.
Big data-based simulation and optimization technology for smart manufacturing
빅데이터기반 제조라인 물류운영 시뮬레이션 및 최적화 기술개발
배터리 생산 공정의 이상상황을 빅데이터 분석 방법을 통해 예측하고 이에 대응 가능한 스케줄링/리스케줄링 알고리즘 작동 프레임워크 제안
by Python
with SKKU, Micube solution, Samsung SDI, KITECH (MOTIE of Korea)
2019.04.~2019.12. (entire 2019.04.~2020.12.)
Development of scheduling theory and algorithms with reinforcement learning for manufacturing systems
강화학습 기반 생산시스템 스케줄링 이론 및 알고리즘 개발
다양한 생산 시스템의 스케줄링을 위한 강화학습 기반 알고리즘 개발 및 성능 검토
by Python (torch)
with SKKU, KAIST (NRF of Korea)
2019.03.~2022.02.
Methodology for dispatching rules' weights
Dispatching rules 가중치 제안
반도체 FAB 생산라인의 디스패칭 룰 기반 자동화 시스템의 최적 파라미터 도출
by C# (MozArt), R
with SKKU, SK Hynix
2018.07.~2018.09.
Framework development for KPI analysis with various weights on dispatching rules
디스패칭 룰 가중치에 따른 KPI 변화 분석 프레임워크 개발
반도체 FAB 생산라인의 디스패칭 룰 기반 자동화 시스템의 최적 파라미터를 도출하기 위한 머신러닝 기반 알고리즘 개발 및 상관성 분석
by C# (MozArt), R
with SKKU, VMS Solutions
2018.07.~2019.02.
Analysis of KPIs according to weights on dispatching rules for LCD manufacturing
LCD 생산 공정 스케줄링을 위한 디스패칭 룰 가중치에 따른 KPI 변화 분석
LCD FAB 생산라인의 디스패칭 룰 기반 자동화 시스템의 파라미터에 따른 스케줄 성능 변화를 체계적으로 분석할 수 있는 프레임워크 개발
by C# (MozArt), R
with SKKU, VMS Solutions
2017.07.~2018.02.
Design and analysis for operations optimizations of smart factory testbed
마이크로 스마트팩토리 테스트베드 가상공장 설계, 운영 분석을 통한 개선 및 최적화 방안 도출
3D 프린터 기반 고객 맞춤 제조 시스템의 유연적인 생산 자동화를 위한 스케줄링/리스케줄링 알고리즘 고도화 및 성능 검토
by Python
with SKKU, ETRI
2017.07.~2018.01.
Development of scheduling and rescheduling algorithms for 3D printer-based smart factory
3D 프린터 기반 스마트팩토리 최적 스케줄링 및 이상상황 대응 리스케줄링 알고리즘 개발
3D 프린터를 활용하는 유연 생산 시스템의 운영 자동화 측면 이슈사항 정의 및 스케줄링/리스케줄링 알고리즘 개발
by Python
with SKKU (NRF of Korea)
2016.11.~2019.10.
Development of algorithms for detecting and improving inefficient schedules in LCD processes
LCD 생산공정의 스케줄링 문제 탐색 및 개선 알고리즘 개발
LCD FAB 생산라인의 기존 공정 스케줄이 주어졌을 때 휴리스틱 알고리즘을 디자인하여 스케줄의 효율성 향상 도모
by C# (MozArt)
with SKKU, VMS Solutions
2016.07.~2017.02.
Development of open FaaS IoT service platform for mass personalization
개인맞춤 생산을 위한 고객-제조-유통 연계 개방형 FaaS IoT 서비스 플랫폼 기술 개발
3D 프린터 기반 고객 맞춤 제조 시스템의 유연적인 생산 자동화를 위한 스케줄링/리스케줄링 알고리즘 개발
by Python
with SKKU, ETRI, Yonsei Univ., HyVISION system, Coever I&T, PartDB, Latis Global Communications (MSIP of Korea)
2016.04.~2018.05. (entire 2015.06.~2018.05.)
Honors and Awards
First Prize - Simulation Challenge - Gating Control in Semiconductor Fabrication - Winter Simulation Conference - San Antonio, USA, Dec. 2023. ($3,000)
Third Prize - AI Competition: Solving Real-world Problems - Integrated Planning Module using AI - Hankook & Company - Daejeon, Korea, Sep. 2023. ($1,500)
Second Prize - Poster Competition: Solving Industry/Social Problems - RL for Resource Leveling in Shipbuilding - KAIST - Daejeon, Korea, Sep. 2022.
Certificate of Appreciation - Successful Project - Workload Balancing of Ship Cargo Production - Samsung Heavy Industries - Korea, May. 2022.
Ph.D. Candidate Research Incentive Support - National Research Foundation of Korea - May. 2022– Apr. 2023. ($15,000)
Third Prize - 2015-2016 PACE RSMS Competition (Second Year): Manufacturing Engineering - General Motors - Cincinnati, USA, Jul. 2016.
Third Prize - 2015-2016 PACE RSMS Competition (Second Year): Customer Insight - General Motors - Cincinnati, USA, Jul. 2016.
Patents
Optimizing method for the manufacturing process [Link]
공정 최적화 방법
반자동 조립 공정에서의 작업자 할당 실시간 최적화 프레임워크 개발
Application date: 2021.05.31
Application number: KR1020210070359
Applicant: KAIST, Yura (entire 12 people)
Authority: Republic of Korea