A paint coating process for which there is little information known about the relationship between the production variables and the output variables was investigated. The goal was to determine the optimum process settings to achieve the target value for the response, coating thickness. Three planned experiments of eight runs each were carried out, the first with four replicates and the other two with two replicates. The observed data were analyzed using a multiple linear regression. Four input factors appeared to be significant and were used to set the production levels in such a way that the desired coating thickness was achieved.