Recognizing Steel Plate Side Edge Shape automatically using Classi?cation and Regression Models

Abstract

In the steel plate production process it is important to minimize the wastage piece produced when cutting a mother steel plate to the size ordered by a customer. In this study, we build classification and regression models to recognize the steel plate side edge shape, if it is curved or not and the amount of curvature. This is done based on time series data collected at the manufacturing line. In addition, this information needs to be presented in a way that enables fast analysis and long-term statistical monitoring. It can then be used to tune the parameters of the manufacturing process so that optimal curvature can be found and the size of the wastage piece can be reduced. The results show that using the classification and linear regression methods, the side edge shape can be recognized reliably and the amount of curvature can be estimated with high accuracy as well.

AI Enhanced Alarm Presentation for Quality Monitoring

“…Control charts, statistical process control (SPC) and automatic process control (APC) are in everyday use for process monitoring and adjustment. Statistical models give another angle to quality monitoring, as they can be used to predict the future outcome of a process, which in its turn enables planning of the process and the production as a whole [7], [8]. Statistical prediction models have been utilized also in smart city related applications; a driving coach system can assist the driver for more fuel-efficient driving by predicting the fuel consumption with city map Figure 1.…”