Anomaly detection is essential for the monitoring and improvement of product quality in manufactu... more Anomaly detection is essential for the monitoring and improvement of product quality in manufacturing processes. In the case of semiconductor manufacturing, where large amounts of time series data from equipment sensors are rapidly accumulated, identifying anomalous signals within this data presents a significant challenge. The data in question is multivariate and of varying lengths, with an often highly imbalanced ratio of normal to abnormal signals. Given the nature of this data, traditional data-driven methods may not be appropriate for its analysis. This paper proposes a novel unsupervised anomaly detection model for the analysis of multivariate time series data. The model utilizes a unique recurrent neural network architecture and a special objective function to detect anomalies. Furthermore, a relevance analysis method is introduced to facilitate the interpretation and analysis of the detected anomalous signals. Our experimental results indicate that this deep anomaly detection model, which summarizes sensor data of different lengths into a low-dimensional latent space, enabling the easy visualization and distinction of anomalous signals, can be applied in real-world semiconductor manufacturing factories and used by on-site engineers for both analysis and execution purposes. INDEX TERMS Anomaly detection, fault detection and diagnosis, multivariate time series data, semiconductor manufacturing, unsupervised learning.
Journal of Chemical Engineering of Japan, Sep 20, 2020
A stochastic multi-objective mathematical model based on material and energy ows under the system... more A stochastic multi-objective mathematical model based on material and energy ows under the system uncertainty of speci c electricity demand-related scraps characteristics and operation conditions was developed to simulate the electric arc furnace steelmaking process and optimize carbon dioxide emission and cost. Considering several energy-saving technologies based on stochastic thermodynamic energy e ciency and electricity price, this paper suggests an optimal cost-saving and carbon-dioxide-emission-reducing strategy. The suggested model provides a trade-o relationship between cost and CO 2 emission. To minimize CO 2 emissions, a tunnel preheater without additional fuel consumption was suggest In contrast, to minimize the cost of utilizing cost-e ective fossil fuels instead of electricity as the system energy requirement, an oxy-fuel burner and shaft furnace type of preheater were proposed. This problem was formulated as a mixed-integer linear programming model.
Uploads
Papers by Youngwook Bin