As a just‐in‐time (JIT) ordering system for multi‐stage production inventory systems, both the kanban system and its alternative, the concurrent ordering system, have been proposed. For both JIT ordering systems, the reactive JIT ordering systems that adjust the buffer size responding to unstable changes in demand were proposed. In addition to the adjustment of buffer size, switching ordering systems between the kanban system and the concurrent ordering system is proposed in order to realize agile control in JIT ordering systems. In the proposed system, the time series data of demand is monitored, and switching the order release systems and adjusting the buffer size are considered when an unstable change in demand is detected with the exponentially weighted moving average charts. The effectiveness of the proposed system is analyzed using simulation experiments under unstable conditions. From these simulation experiments, it is clear that the proposed system can react to unstable changes in demand and satisfy the required level for the mean waiting time of demand. Also, to attain the required level, the proposed system must possess much less work‐in‐process inventories than the previous systems which adjust only the buffer size and do not switch ordering systems.
Deals with the kanban system in just‐in‐time (JIT) production as a system of production planning and inventory control for multi‐stage production inventory systems, Constructs a queueing network model of the kanban system with SLAM II. Aims to improve the kanban system by modifying the flow of information. Because the orders for all stages are released concurrently, the modified system is called the concurrent ordering system. By means of simulation experiments, investigates and compares the performances of the models.