The simultaneous analysis of thermal variables with normal distribution with the aim of checking if there is any significative correlation among them or if there is the possibility of making predictions of the values of some of them based on othersâ values is considered a problem of great importance in statistics studies. The aim of this paper is to study the applicability of linear regression models in working environmentsâ thermal comfort studies, thus contributing for the comprehension of the possible environmental cooling, heating or winding needs. It starts with a bibliographical research, followed by a field research, data collection and and software statistical-mathematical data treatment. It was then performed data analysis and the construction of the regression linear models using the t and F tests for determining the consistency of the models and their parameters, as well as the building of conclusions based on the information obtained and on the significance of the mathematical models built.