Logo image
Supply chain planning for back-end production networks with mixed traditional and wafer-level chip-scale packaging for IC design houses
期刊文章   同儕審查

Supply chain planning for back-end production networks with mixed traditional and wafer-level chip-scale packaging for IC design houses

James C. Chen, Tzu-Li Chen, Yin-Yann Chen, Chung-Wei Su 和 Dewanti Anggrahini
JOURNAL OF THE CHINESE INSTITUTE OF ENGINEERS, 頁碼.1-33
26/08/2026
Web of Science ID: WOS:001860909700001

摘要

Engineering, Multidisciplinary Science & Technology Engineering Technology
The importance of supply chain planning (SCP) has increased in the semiconductor industry, as production planning involves multiple fabrication sites across a multi-site supply chain network. Allocating materials and capacity within the supply chain to maximize profit has become a critical challenge. This research aims to propose a back-end production SCP model for integrated circuit (IC) design houses that covers testing, assembly, and packaging. The proposed model considers both traditional package and wafer-level chip-scale package production processes, accounting for additional routes and technical constraints. Planning in the back-end production is driven by a combination of deterministic additional-demand forecasts and actual customer-confirmed orders to determine production, inventory, and transportation quantities for each process stage. The objective is to maximize total profit, subject to constraints on machines and the flexibility of substitutes. The effectiveness and significance of the proposed mixed-integer linear programming (MILP) SCP model were evaluated through experiments that accounted for multiple factors. Due to the NP-hardness of the MILP model, two decomposition-based algorithms were proposed to solve it, achieving reasonable solution quality and significantly faster computation times than other methods. Additionally, a sensitivity analysis was conducted to assess the impact of different parameters on production costs and total profit.

相關連結

指標

1 檢視次數

詳細資料

Logo image