SCM 환경에서 행동-보상 학습 기법을 이용한 JIT 기반 동적 간판 시스템

Abstract

This paper propose a reactive Kanban system for a multistage production system with non-stationary customer demands using action-reward learning method. The proposed method leads to decrease the inventory cost. In the reactive Kanban system, the time series data for the demand are monitored and the control parameter of action-reward learning is designed to adaptively change as customer demand pattern changes. Then, in this system, the number of Kanbans and the buffer size at each stage are adjusted as a response to the adapted control parameter. A simulation-based experiment was performed to compare the performance of the reactive Kanban system.