In this paper a novel approach for monitoring tool-related faults in milling processes by utilizing process simulation-based machine learning algorithms, specifically Random Forest algorithms, for fault detection is presented. In order to train machine learning models in tool condition monitoring, laboratory tests have traditionally been required. This method eliminates the need for costly, time-consuming laboratory tests. The training process has been simplified by utilizing analytical simulation data and provides a more cost-effective solution by leveraging analytical simulation data. Based on the results of this study, the proposed approach has been demonstrated to be 94% accurate at predicting tool-related faults, demonstrating its potential to serve as an efficient and viable alternative to conventional methods. These findings have been supported by actual measurement data, with a notable accuracy rate of 93% in the predictions. Furthermore, the results indicate that process simulation-based machine learning algorithms will have a significant impact on the tools condition monitoring and the efficiency of manufacturing processes more generally. To further enhance the capabilities of the proposed fault monitoring system, process-related and machine-related faults will be investigated in future research. Several machine learning algorithms will be explored as well as additional data sources will be integrated in order to enhance the accuracy and reliability of fault detection.