© 2025 by the authors.Featured Application: We propose a reference framework designed to foster model deployment in an industrial context by promoting the self-monitoring and updating of models when required. The impact on decision-making processes is explored to enable reinforcement learning in the context of this framework. The aim of this study was to evaluate machine learning algorithms’ capacity to improve prescriptive maintenance. A pumping system consisting of two hydraulic pumps with an electric motor from a Spanish petrochemical company was used as a case study. Sensors were used to record data on the variables, with the target variable being the bearing temperature of the electric motor. Several regression models and a neural network time series model were tested to model the system variables. A bearing temperature sensitivity analysis was conducted based on the coefficients obtained from the optimization of the regression model. To fully exploit the capabilities of these techniques for application in this field, we designed a reference framework intended to foster model deployment in an industrial context by promoting the self-monitoring and updating of the models when required. The impact on decision-making processes is explored using reinforcement learning in the context of this framework.