Application of the bootstrap method for statistical characteristics assessment of aircraft components’ small samples

The estimation of adequate service life of aircraft instruments is a factor of great importance in aircraft operation process. Changing the instrument service interval affects both reliability (shorter intervals make it easier to locate malfunctions of components and assemblies as early as possible) and economic performance (inducing increase of operating costs). So, the increasing the service interval without potentially reducing reliability is an economically im- portant task. To determine the optimal time to maintenance for aviation components and assemblies, it is necessary to determine the span of their service life with the highest degree of precision. The problem of calculating such estimates is compli- cated by the fact that the data on component failures are scattered and incomplete, which makes it difficult to assess their statistical characteristics accurately. The purpose of this article is to find an effective method of statistical characteristics assessment for small samples as the first stage of modeling of the aircraft components and assemblies reliability. It is induced by specific operational factors of aviation components exchange at small airlines operating Soviet-time aircraft. The article examines two methods of resampling, bootstrap and jackknife. There is also an assessment of mean time to failure expectation for fuel gauges, of the variance and root-mean- square deviation in the article. The bootstrap method is offered as applicable for statistical characteristics assessment of mean time to failure ex- pectation for aircraft components and assemblies taken for analysis in small samples (pressure gauges were chosen as an example of such analysis). The assessments and calculations can be used by airlines to state the nonfailure service time of a variety of components and assemblies.