• Proceedings of the Korean Society for Quality Management Conference
• /
• 2006.11a
• /
• pp.221-226
• /
• 2006

We usually use two techniques(influence function and local influence) for detecting outliers. But, we cannot use these difficult techniques in elementary industrial statistics course for college students. We can use some simple graphical methods(box plot, dandelion seed plot, influence graph and cumulative deletion plot) for univariate and bivariate outlier detection and outlier effect in elementary industrial statistics course for college students.

• Journal of Conservation Science
• /
• v.33 no.2
• /
• pp.75-83
• /
• 2017

This study produced smelting slag through the reproduction of an ancient iron manufacturing technique, with the aim of facilitating a comprehensive understanding of the process by analyzing the slag components. The research suggests an interpretation method using the ratio of the subcomponents relative to the main slag components as an alternative to existing methods. We investigated the component source within the smelting furnace from which the slag is derived by developing an understanding of the tendency between slags. Based on bivariate graph and triangular coordinate data analysis, it was found that a slag can be categorized according to its components. The groups were identified as the ore slag group(centered on the ore), and the clay slag group(centered on clay and granite soil). This research determined that it is possible to estimate the components derived from the slag, depending on which group they belong to or resemble, as shown in Figure 4~7. It was found that a comprehensive understanding of the ratio between the components was more accurate than a simple analysis of the contents, for the interpretation of ancient iron manufacturing processes. This is based on the fact that a higher ratio of $TiO_2$ was detected by the components analysis, and an analysis of all the slag showed that the value of $CaO/SiO_2$ ratio was lower than 0.4, which corresponds to the reproduction experiment condition in which flux was not used.