3. Production Process Characterization 3.4. Data Analysis for PPC 3.4.4. Can I tell which factors are causing the most variation in my process?
Studying variation is important in PPC.	One of the most common activities in process characterization is to study the variation associated with the process and try to determine the important sources of that variation. This is called analysis of variance. Refer to the section of this chapter on  ANOVA models for a discussion of the theory behind this kind of analysis.
The key is to know the structure.	The key to performing an analysis of variance is identifying the structure represented by the data. In the ANOVA models section we discussed one-way layouts and two-way layouts where the factors are either crossed or nested. Review these sections if you want to learn more about ANOVA structural layouts.
	To perform the analysis, we just identify the structure, enter the data for each of the factors and levels into a statistical analysis program and interpret the output ANOVA table. This is all illustrated in the example below.
Example: furnace oxide thickness with a 1-way layout.	The example below is a furnace operation in semiconductor manufacture where we are growing an oxide layer on a wafer. Each lot of wafers are placed on quartz containers (boats) and then place in a long tube-furnace. They are then raised to temperature and held for a period of time in a gas flow. We want to understand the important factors in this operation. The furnace is broken down into four sections (zones) and two wafers from each lot in each zone are measured for the thickness of the oxide layer.
	The first thing to look at is the effect of zone location on the oxide thickness. This is a classic  one-way layout. The factor is furnace zone and we have four levels. A plot of the data and an ANOVA table are given below.
The zone effect is masked by the lot to lot variation.
	Analysis of Variance     Source DF SS Mean Square F Ratio Prob > F zone 3 912.6905 304.23 0.467612 0.70527 Within 164 106699.1 650.604
Let's account for lot with a nested layout.	From the graph, there does not appear to be much of a zone effect, in fact the ANOVA table indicates that it is not significant. The problem is that variation due to lots is so large that it is masking the zone effect. We can fix this by adding a factor for lot. By treating this as a nested two-way layout, we get the ANOVA table below.
Now both lot and zone are revealed as important.	Analysis of Variance     Source DF SS Mean Square F Ratio Prob > F lot 20 61442.29 3072.11 5.37404 1.39e-7 zone[lot] 63 36014.5 571.659 4.72864 3.9e-11 Within 84 10155 120.893
	Sine the "Prob > F" is less than .05, for both lot and zone, then we know at the 95% level of confidence that these factors are statistically significant.