On the Optimal Replenishment Models and Strategies for Multi-Stage Production Supply Chain Systems

• Main Authors: Tien-Cheng Hsu, 許天成
• Other Authors: Ming-Jong Yao
• Format: Others
• Language: zh-TW
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/55650961142480418157

博士 === 東海大學 === 工業工程與經營資訊學系 === 97 === We are interested in the optimal replenishment models and strategies in complex multi-stage production supply chains in this study. Nowadays, supply chain systems encounter more intensive competition and their operations become much more complicated as they face the trend of globalization. Though efficient control of the operations is critical for a multi-stage supply chain in practice, we found that theoretical studies are difficult and limited number of research articles were published in this domain in the literature. In this study, we focus ourselves on the deterministic production planning in serial-type supply chain systems and its extension, namely, assembly-type supply chain systems. We would like to determine the optimal production period, the production lot and the shipment batches so as to minimize the average total cost incurred in the supply chain systems. This study explores two decision-making scenarios in multi-stage supply chain systems as follows. First, kanban, the de facto mechanism in a Toyota Production System production line, is incorporated into multi-stage production supply chain systems, where Just-In-Time philosophy is applied. We are interested in the coordination of the production stages by a common optimal production period and the determination of the optimal number of kanbans in each production stage. Second, we would like to investigate the Joint Economic Lot-Sizing Problems, where shipments are multiple, shipment batches can be delivered before production lot is finished, and batch sizes could be unequal in a production period. (Especially, geometric series batch sizes are interested.) For both scenarios, we conduct theoretical analysis that provides interesting insights into the optimality structure of the mathematical model. Also, by making use of our theoretical results, we propose effective algorithms that guarantee obtaining a global optimal solution for the corresponding scenario. After verifying with numerical experiments, we assert that the proposed solution methodology shall serve as an effective decision-support tool that assists the managers to determine optimal replenishment strategies in multi-stage production supply chain systems.