Prediction of the reliability of non-repairable elements directed to street lighting lamps of street lighting

Abstract

The purpose of this study is to determine the necessary steps for the calculation of reliability indicators in non-repairable equipment, the study was carried out on a sample of 250 100-watt sodium lamps for public lighting. Being the time to failure TTF the most important parameter to perform a reliability analysis on non-repairable equipment, (time since the equipment installation until its failure), in this case the equipment cannot be repaired, therefore, it must be replaced. After the failure, the optimal data collection is essential to carry out a correct analysis of the reliability indicators. It is necessary to find which of the parametric functions is the one that fits the best the collected data, for this it is imperative to find the Function F(t) that represents the parametric distribution accumulated time to failure and compare it with the non-parametric function or empirical function. Parameters of all selected hypothetical distributions must be calculated with the data collected in this study. For each distribution, the test value is calculated, which is compared with the critical value of the Kolmogorov-Smirnov kindness fit test, therefor defining whether the hypotheses are rejected or not. If rejected, it means that the value of the test is greater than the critical value, otherwise the hypotheses are accepted, and the lowest value decides the parametric distribution that fits the best with the sample data. Once the distribution has been selected, reliability indicators can be estimated, such as the cumulative failure probability function F(t), the Survival Function R(t), and the failure rate function, also known as the risk function λ(t). The method will provide a very useful tool for calculating the reliability of repairable equipment and its future evaluation, which will provide ideas for decision-making based on continuous improvement.