Reliability Analysis of Military Vehicles Based on Censored Failures Data

The paper proposes a methodology of reliability testing as applied to vehicles used in military transport systems. After estimating the value function using Kaplan–Meier estimator, models were developed and analysed. neural model, which achieved correlation coefficient R exceeding 0.99, was determined fit empirical data best. On basis approximated several models, characteristics tested sample determined. Plots failure probability density for all three had similar courses over significant part domain. A intensity also determined, varied between models. For exponential Weibull expected mileage failures calculated, proved impossible model. proposed is capable modelling based on observation an assumed period exploitation process selected group vehicles.