國 立 交 通 大 學 演 講
講 題：Stochastic Modeling on Supply Chain Originated from Real-World Cases.
主 講 人：張國華 教授 (中原工工)
主 持 人：陳文智 教授
時 間： 107 年 12 月 3 日 (星期一)13:20~15:20
地 點：管二館520 室
|1. Some refurbished products are still useable and popular. This brings up a new business model collecting the used or failure products for refurbishing and providing them to customers when customers bring back their failure products and some emerging company issues insurance policy or extended warranty to the high-end products such as cellphones. We model the corresponding supply chain system as a queueing model on which we determine the important control parameters such as the base stock level, time to place order, the corresponding order quantity and time to sell back.
2. In production system, the quality of the semi-product is certainly affected by machine working condition and it may also deteriorate as it waits in the production process. We define the framework of the corresponding queueing system. By utilizing absorbing time analysis, we estimate the final quality of the current product given the state of the system and we estimate the steady-state distribution of final quality and hence the yield of the end products.
3. The aviation and defense large scale product systems such as fighter, battleship and radar have to fulfill the required availability and it needs sufficient quantity of spare parts to restore the system promptly. We first define the tree structure for the maintenance system and define the number of functional leaf parts at the bottom layer as our states and set up the relation between the number of functional modules in other layers, based on which we can describe the whole system. By applying continuous time Markov chain, we obtain the steady-state probabilities of state and, from which, we obtain the availability of the system product and the fill rate of each module or part.