193 – Contributed Oral Poster Presentations: Section on Statistical Education
Variation Exploration Lab in a Manufacturing Setting
Diane Evans
Rose-Hulman Institute of Technology
In Engineering Statistics, a great deal of attention is paid to variation in a process. In every process, there is variation. Statistical process control charts were developed to separate variation in a process into two components-common cause and special cause. The type that is inherent in a process is called common cause variation; it is there no matter what we do. Variation that drastically affects a process is called special cause variation, and is usually due to just that-special cause. A process is said to be "out of control" if the data plotted on a control chart shows patterns, trends, or unusual jumps. Otherwise, we say the process is "in control." How we sample and group our data to construct a control chart is really dependent on what we want our chart to tell us. For example, control charts can be used to determine if there is variation between our subgroups or even within a subgroup. I have developed a classroom lab in which students work at a company that manufactures Easter peeps. In order to provide answers to their main stakeholders about how "well" the process is running, students must figure out how best to group their data to tell the appropriate "story."