In the industrial context, thousands of data are generated every day, which, once processed, can be transformed into valuable information for the business and to support decision-making. This paper is applied to the case study of the textile industry - one of Portugal's oldest and most important industries, which has been reorganizing itself over the years to keep up with changes in the market. However, regarding data exploitation, decisions are primarily based on opinions and assumptions, not based on the daily data. This study aims to apply a set of data mining techniques to solve the classification problem in the textile industry: predicting, when planning production, whether an order will be delayed. In the study, 28 models were included. These models are designed based on business knowledge, considering variables that can influence the target. Overall, this study shows the potential of applying these models in an industrial context and using them to help plan orders. The best model developed has an accuracy of 94%, a sensitivity of 90%, a specificity of 98%, a precision of 97% and an F-Measure of 93%.