Search for: All records

Smart manufacturing systems are considered the next generation of manufacturing applications. One important goal of the smart manufacturing system is to rapidly detect and anticipate failures to reduce maintenance cost and minimize machine downtime. This often boils down to detecting anomalies within the sensor data acquired from the system which has different characteristics with respect to the operating point of the environment or machines, such as, the RPM of the motor. In this paper, we analyze four datasets from sensors deployed in manufacturing testbeds. We detect the level of defect for each sensor data leveraging deep learning techniques. We also evaluate the performance of several traditional and ML-based forecasting models for predicting the time series of sensor data. We show that careful selection of training data by aggregating multiple predictive RPM values is beneficial. Then, considering the sparse data from one kind of sensor, we perform transfer learning from a high data rate sensor to perform defect type classification. We release our manufacturing database corpus (4 datasets) and codes for anomaly detection and defect type classification for the community to build on it. Taken together, we show that predictive failure classification can be achieved, paving the way for predictive maintenance.

Monitoring product temperature during lyophilization is critical, especially during the process development stage, as the final product may be jeopardized if its process temperature exceeds a threshold value. Also, in-situ temperature monitoring of the product gives the capability of creating an optimized closed-loop lyophilization process. While conventional thermocouples can track product temperature, they are invasive, limited to a single-point measurement, and can significantly alter the freezing and drying behavior of the product in the monitored vial. This work has developed a new methodology that combines non-invasive temperature monitoring and comprehensive modeling. It allows the accurate reconstruction of the complete temperature profile of the product inside the vial during the lyophilization process. The proposed methodology is experimentally validated by combining the sensors’ wirelessly collected data with the advanced multiphysics simulations. The flexible wireless multi-point temperature sensing probe is produced using micro-manufacturing techniques and attached outside the vial, allowing for accurate extraction of the product temperature.