<p>In semiconductor assembly manufacturing, the storage and transportation of wafers between processing stages play a critical role in maintaining product quality and ensuring stable production flow. After completing certain sensitive operations, wafers may become vulnerable to environmental exposure such as oxygen and moisture during waiting periods. When downstream equipment cannot receive wafers within an acceptable time window, temporarily storing wafers in environmentally controlled facilities such as nitrogen cabinets becomes necessary to prevent quality degradation. However, coordinating wafer transportation and intermediate storage decisions in automated material handling systems (AMHS) remains a challenging operational problem in semiconductor assembly systems. Hence, this study develops an integrated optimization model that jointly considers wafer routing, machine assignment, and nitrogen cabinet utilization. The model captures conditional transfer behavior in which wafers may either be directly transported to downstream machines or temporarily stored in nitrogen cabinets depending on system conditions. To efficiently solve the resulting problem, a hybrid elite genetic algorithm with variable neighborhood search (EGAVNS) is proposed to efficiently obtain high-quality solutions. Extensive computational experiments conducted under different problem scales demonstrate that the proposed approach significantly improves solution quality and convergence stability compared with other algorithms. The proposed algorithm provides an effective decision-support tool for semiconductor assembly manufacturing by improving the coordination between wafer storage and transportation, thereby enhancing production efficiency while maintaining wafer quality during inter-process waiting periods.</p>