Response surface methods were applied to a computational missile model to investigate their ability to adequately predict simulation results, identify influential parameters, and decrease the amount of time needed for missile analysis. Three separate studies were conducted based on different input combinations of propellant mass flow rate, specific impulse, drag coefficient scale factor, pitch over rate gain, and launch elevation angle. A simulation matrix was designed for each study and executed to produce a polynomial response equation for the missile range. The residual range prediction errors of the response equation were computed over the range of input parameters. The response surface methods were found to adequately predict simulation results, identify the influential input parameters, and significantly reduce the analysis time. |