Viscosity analysis in lubricants through factorial design with three factors

Abstract

In the field of statistics, the experimental design becomes an analysis tool to establish the ideal conditions of a very important process and / or product, analysis of variance it is very useful in cases where there are a series of controllable factors and that affect a complex process in order to optimize them, in the specific case, it is sought to give a solution to a lubricants analysis using a factorial design. For the development of the following document, it is proposed to show the analysis performed on lubricant samples by means of a 3 x 22 factorial design with 9 replicas. The analysis was performed on 108 samples, taking into account three different types of lubricants; which were subjected to a temperature of 40 degrees Celsius by kinematic viscocity test. It was observed in the analysis that taking into account the factors that interacted in the design (types of lubricant, parts, brands), only one of them showed significant interaction. The use of the design of factorial experiments is considered pertinent, since this can be useful in making maintenance decisions when buying a specific lubricant brand. The experiment design was carried out with StatGraphics statistical software. © 2018 Latin American and Caribbean Consortium of Engineering Institutions. All rights reserved.