Analysis of predictive factors for fully mission capable rates of deployed aircraft

Abstract

As the U.S. military reduces its forces, the ability to maintain an acceptable level of readiness is of concern to the U.S. Navy. Both personnel and equipment readiness and the ability to predict them have been the focus of much attention. Fully Mission Capable (FMC) rates measure the percentage of time that aircraft are fully able to meet mission requirements. FMC rates have been determined to be the best single measure of equipment condition, providing an indication of aircraft readiness. This thesis evaluates the capabilities of logistic regression and regression trees in predicting aircraft readiness for a specific carrier deployment or aircraft type/model/series (TMS). The data are taken from observations of squadrons by aircraft TMS by month from 1981 through 1997. Empirical results indicate that logistic regression and regression trees provide forecasting results with standard errors of prediction better than taking the mean and standard deviation of the historical data.