Date of Award


Publication Type

Master Thesis

Degree Name



Mechanical, Automotive, and Materials Engineering

First Advisor

Stoilov, Vesselin

Second Advisor

Sobiesiak, Andrzej


Fourier, Robustness, Sensitivity, Simulation, Thermal, Uncertainty




This study has two purposes: (1) to investigate the effects of uncertainty on thermal design and performance (2) to establish a process to couple the results from uncertainty analysis with Design for Six Sigma (DFSS) to improve robustness. The research was conducted with support from FCA group. The analysis was applied to a component selected by FCA group; however, it is applicable to any system. The effects of uncertainty were investigated using the Fourier Amplitude Sensitivity Test (FAST). The analysis was run on MATLAB software package using estimated uncertainty ranges for each parameter. The parameters with the greatest influence on thermal performance uncertainty were identified. These parameters were used as control factors in a DFSS study. The thermal performance was simulated on RadTherm software package with several settings for the critical parameters. The settings which allowed the performance to remain consistent, regardless of noise, were determined. Thus, robustness was improved.