Track: Undergraduate Student Paper Competition

Abstract

The purpose of this study is to describe a model of a practical experiment that can be applied in the classroom about the importance of concepts associated with the Design of Experiments course using the R language and SPSS. The completely randomized block design of the experiment model was selected in order to determine whether or not there are significant differences in the fragrance durations of four types of perfumes with four tasters, of the NATURA perfume brand. Then, in order to ensure the accuracy of the results, the normality, homoscedasticity of variances and independence of residuals tests were verified. After performing the experiment, the ANOVA analysis indicated that there is a significant difference in the fragrance durations of the perfume types at 95% confidence and also evidence that there is no significant difference in the sensory evaluation of the tasters. We then proceeded to perform a Post Hoc analysis: DMS, Tukey and Bonferroni with α=0.05, which helped to determine the pairs of means different from each other, detecting the type of perfume with the longest fragrance duration. The study brings the importance of understanding that the subjectivity of the tasters have no influence on the duration of the fragrances and the improvement of the perfume quality or the development of new fragrances in the market.