Abstract:

This research addresses the optimization of capacity allocation in the photolithography area in a semiconductor manufacturing company using Linear Integer Programming (LIP). The primary objective of this research is to optimize machine utilization rates and balance workloads across all machines while meeting wafer demand. To address this problem, we aim to develop an LIP model that optimizes capacity allocation and determines the optimum wafer allocation quantities for each machine and recipe. The study begins with a comprehensive explanation of semiconductor manufacturing, emphasizing its significance and challenges in the photolithography area. The methodology includes problem identification, data collection, analysis, LIP model development to be applied using MATLAB, and results validation through sensitivity analysis.  Results show that the optimization improved the average machine utilization from 88.87% to 69.46% and reduced the Mean Absolute Deviation (MAD) from 43.23 to 0.003, indicating a more balanced workload. The study concludes with key findings, contributions to semiconductor manufacturing field, and directions for future research.