Lube Based Oil Unit Anomaly Limit Prediction using Decision Tree Classification Model

The replication of human thinking by technology, including computer systems, is known as artificial intelligence (AI). The rise of AI can certainly help plant operations. In the past, process engineers manually predicted anomalous conditions in the plant to ensure every process was dependable and safe. This extensive procedure calls for an experienced and skilled engineer to find anomalies. This makes the job more difficult for engineers at the entry-level. This pattern is observed throughout the industry, where process plants with newly hired engineers tend to have more accidents and unit upsets because the engineers fail to identify anomalies early on. Thanks to the development of artificial intelligence and machine learning, we can now use all this expertise to detect early normal on computers. We can now use all this expertise to detect early normal data on computers due to the development of artificial intelligence and machine learning from historical data. For this study, we have gathered data from plant engineers over the last five years on common or prior events that have happened in the past. A decision tree learning model has been trained with the knowledge gathered from five fault tree analyses. A high degree of accuracy (99.9%) was achieved using supervised machine learning training when the model was evaluated using several data sets. Now that it can be used in production, this model helps engineers find anomalies early on.