This study investigates the applicability of the Random Forest model in conjunction with Big Data for analyzing machine operating data, with the aim of improving the effectiveness of predictive maintenance and mitigating the associated operating costs. The introduction outlines the vital importance of Big Data and Machine Learning in analyzing voluminous data sets in order to extract information of strategic relevance. The aim of the study is to evaluate machine operating data using the Random Forest technique, with a focus on reducing operating costs through predictive maintenance strategies. The literature review contextualizes the relevance of Big Data and modelling techniques in generating insights in various domains, including the prognosis of equipment failures. The methodology describes in detail the procedures adopted for data collection and analysis, using computer simulation and data processing tools in Python. The results show that operational overload is the main precursor to failures, culminating in high temperatures and premature tool wear, resulting in unplanned downtime and significant operating costs. It is concluded that implementing strategies such as predictive maintenance supported by classification models can not only optimize operational efficiency, but also reduce waste and maximize return on investment, providing valuable perspectives for managing large volumes of data in a variety of business contexts.