The Case of the Variable Laminates Granting University Experimental design is “a formal plan that details the specifics for conducting an experiment, such as which responses, factors, levels, blocks, treatments, and tools are to be used” (Shower 2010). An experiment where one variable is studied while the other variables are held constant can be inefficient and suffers from the inability to assess interactions among the variables as well as the interaction effects among the variables (Shower 2010).
An experimental design that I think will assist the process engineers in determining what actions to take to reduce laminate thickness variation is the Full-Factorial Design. There are multiple factors to be considered when trying to reduce the laminate thickness variation. A full-factorial allows for all possible combinations of the factors in an experiment at a number of levels to be tested. It allows for the estimation of all main effects and all interaction effects.
The process engineers should take a look at the temperature, erasure, and the time that is being used for each process. The specification for soaking the log calls for 60 minutes but there is no process control in place (Shower 2010). The specification for the pressure of the knife is between 250 SSI and 300 SSI therefore the specification for the knife pressure is not standard. The soak temperature for the logs varies between 150 degrees and 200 degrees, depending on how much bark has accumulated around the heating coils.
By using the 23 factorial designs, the process engineer can examine the three levels of the three factors which will show complete evidence of what is causing the variation in the thickness. This will cause the process to have more runs than using the Single-Factor or the One-Factor-at- a-Time Design but it will be more accurate in determining the reason behind the laminate thickness variation.