• ANOVA
• confidence interval analysis
• discrete event simulation
• full factorial DOE
• manufacturing system
• Robustness

Uncertainty affects significantly manufacturing systems both within the plant boundaries and externally. Therefore, many companies simulate their production processes to cope with disturbances and evaluate robustness. Collaborating with an aerospace manufacturing firm, the scope of this paper is to devise a methodology to evaluate the robustness of a manufacturing system, starting from an already built discrete event simulation (DES) model in Tecnomatix Plant Simulation software. Different scenarios to test the simulation model were established and the confidence interval method was applied to assess key performance indicators (KPI) robustness against different disturbances. To analyse system variability two different ANOVA were conducted. The first one, between scenarios, evaluated the disturbances effects onto the system, while the latter, within each scenario, compared the dispatching rules. A design of experiment analysis was performed to assess disturbances interaction. Finally, a cost model was defined to perform an in-depth comparison between the policies. The analysis pointed out the minimum number of observations to get a robust system in the different operating conditions, the factor with the greatest impact on performances and the best policy to face disturbances allowing a fitted performances improvement.