1. Aircraft Windshield Failures Statistical Methods for Reliability Engineering Professor Gutierrez-Miravete Erica Siegel December 4, 2008

2. 2 Problem Statement Objective: Aircraft windshield failure data will be analyzed to determine the best-fit distribution. System: Windshield is comprised several layers of material all laminated under temperature and pressure. Failures refers to the delamination of the outer layer of the windshield or failure of heating system, not a structural defect. Although the windshield itself is not damaged it must be replaced. Management issue due to the potential cost of different warranties and warranty periods. Data was taken from the text Reliability: Modeling, Predictions and Optimization.

3. 3 Methodologies First, Minitab was used to determine the best fitting distribution. Weibull, with a correlation coefficient of.975, was chosen as the fit. The shape and scale coefficients were then determined. Mapel was then used to manipulate the Weibull equation to find the reliability function (R.), the probability density function (f), the hazard rate function (z), and the mean time to failure (MTTF). Finally, a Monte Carlo simulation was run in excel to verify the results from Maple. Note: All data in 10000 hours

4. 4 Conclusions The mean time to failure (MTTF) of the system is 2615 hours. Monte Carlo simulation is a good predicator of failure rates for this data. Large sample size is critical to finding a well-fitting distribution. Prediction equations are very useful in determining system failure rates.