Maintenance is the technology of continuously monitoring the conditions of equipment and predicting the timing of maintenance for equipment. Particularly in the field of semiconductor manufacturing, where processes are automated, various methods are being tried to minimize the economic losses and maintenance costs caused by equipment failure. A new Predictive Maintenance (PdM) technique, a new method of maintenance, is introduced in this paper to develop an algorithm for predicting the failure of wafer transfer robots in advance. The acceleration sensor data used in the experiment were obtained by installing a sensor onto the wafer transfer robot. To analyze these data, the data preprocessing and FFT process were performed. These data were divided into normal data, first error data, second error data, and third error data (failure data) in stages. By clustering the data using the K-means algorithm, the center point distribution of the clusters was analyzed, and the features of the error data and normal data were extracted. Using these features, an artificial neural network model was designed to predict the point of failure of the robot. Previous research on maintenance systems of the transfer robot used fewer than 50 error data, but 1686 error data were used in this experiment. The reliability of the model is improved by randomly selecting data from a total of 2248 data sets. In addition, it was confirmed that it was possible to classify normal data and error data with an accuracy of 97% and to predict equipment failure by applying neural network modeling.

High-definition (HD) maps determine the location of the vehicle under limited visibility based on the location information of safety signs detected by sensors. If a safety sign disappears or changes, incorrect information may be obtained. Thus, map data must be updated daily to prevent accidents. This study proposes a map update system (MUS) framework that maps objects detected by a road map detection system and the object present in the HD map. Based on traffic safety signs notified by the Korean National Police Agency, 151 types of objects, including traffic signs, traffic lights, and road markings, were annotated manually and semi-automatically. Approximately 3,000,000 annotations were trained based on the you only look once (YOLO) model, suitable for real-time detection by grouping safety signs with similar properties. The object coordinates were then extracted from the mobile mapping system point cloud, and the detection location accuracy was verified by comparing and evaluating the center point of the object detected in the MUS. The performance of the groups with and without specified properties was compared and their effectiveness was verified based on the dataset configuration. A model trained with a Korean road traffic dataset on our testbed achieved a group model of 95% mAP and no group model of 70.9% mAP.

Faults in the wafer transfer robots (WTRs) used in semiconductor manufacturing processes can significantly affect productivity. This study defines high-risk components such as bearing motors, ball screws, timing belts, robot hands, and end effectors, and generates fault data for each component based on Fluke’s law. A stacking classifier was applied for fault prediction and severity classification, and logistic regression was used to identify fault components. Additionally, to analyze the frequency bands affecting each failed component and assess the severity of faults involving two mixed components, a hybrid explainable artificial intelligence (XAI) model combining Shapley additive explanations (SHAP) and local interpretable model-agnostic explanations (LIME) was employed to inform the user about the component causing the fault. This approach demonstrated a high prediction accuracy of 95%, and its integration into real-time monitoring systems is expected to reduce maintenance costs, decrease equipment downtime, and ultimately improve productivity.