This study shows how different modeling assumptions made in discrete-event simulation models can change predictions of maintenance costs and power losses for floating offshore wind farms. By testing two models under the same conditions, we identify which assumptions matter most and how they shape results. The findings help improve the reliability of future models and support better planning of maintenance strategies for floating-wind projects.