An agent-based system for abnormal flow detection in semiconductor production line
Abstract
In this paper, we propose an agent-based system to detect abnormal process flow in semiconductor manufacturing process. A large-scale automation system like a semiconductor line is composed of multi-agents. Each agent carries out only a given role with autonomy. Therefore, even if an abnormality occurs, it might cause huge accident easily. For quality control, it is necessary to detect such anomalies promptly and take necessary measures, but the system complexity is so high that process managers are unable to detect abnormalities. In order to allow the administrator to quickly recognize the problem situation, we propose an agent that monitors process flow, and detects abnormal flow, and verify its validity using the data.
Authors: Dong-Chang Lee, Sung-Bae Cho
Published in: International Conference on Control, Automation and Systems (ICCAS) (2017)