{"id":21374,"date":"2025-11-03T11:50:03","date_gmt":"2025-11-03T10:50:03","guid":{"rendered":"https:\/\/e-opex.pl\/?p=21374"},"modified":"2026-01-14T16:42:22","modified_gmt":"2026-01-14T15:42:22","slug":"design-of-experiments-doe-what-to-measure-during-an-experiment-and-what-experimental-design-to-choose-part-2","status":"publish","type":"post","link":"https:\/\/e-opex.pl\/en\/design-of-experiments-doe-what-to-measure-during-an-experiment-and-what-experimental-design-to-choose-part-2\/","title":{"rendered":"Design of Experiments (DoE). What to measure during an experiment and what experimental design to choose, Part 2\/6"},"content":{"rendered":"\t\t
\n\t\t\t\t
\t\t\t
\r\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

The first part of the article in which we established the goals of the experiment, selected factors and their testing levels,, can be found here<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t
\r\n \r\n <\/i> \r\n PART 1 <\/span>\r\n <\/a>\r\n<\/div>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\n
\n\t
\n\t\t\n\t\t

\n\t\t\t\t\t\t\tWhat will we measure during DOE?\t\n\t\t\t\t\n\t\t<\/h2>\n\t\t\n\t<\/div>\n<\/div>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\n
\n\t
\n\t\t\n\t\t

\n\t\t\t\t\t\t\t1. Y, or preferably the main Ys, are numerical metrics describing the effect of the tested factors\t\n\t\t\t\t\n\t\t<\/h3>\n\t\t\n\t<\/div>\n<\/div>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

We therefore need to measure the yolk quality and ease of peeling. At this point, we assess this only visually, at most commenting on the cooking results. For DOE, we need continuous data. If all we know after the test is whether the egg is good or bad, analysis will be practically impossible. If, for example, our current process only involves visual assessment, we need to find a numerical metric for DOE. There are several ways to do this, but more on that another time.<\/p>

How will we manage this DOE? We will use Adobe Capture to assess the color. This application will give us a color reading for each egg (after cutting it) in the additive RGB (Reed Green Blue) color model, hence:<\/p>

Y1 \u2013 Yolk Goldenness [Reed\/Green].<\/strong> The R\/G ratio is an indicator of the color shift toward warmth. When R is close to G, we have a neutral yellow color. When R<G, the color shifts towards green, meaning a less appetizing yolk. As R>G increases, the color becomes warmer, more golden, and sometimes slightly orange.<\/p>

Y2 \u2013 Green Ring<\/strong> \u2013 the G channel value indicates an overcooked egg. The higher the G level, the less appetizing the egg looks.<\/p>

The next Y will be the comparative scale.<\/p>

Y3 \u2013 Visual Yolk Quality<\/strong> \u2013 a scale created after the doe, by arranging the eggs from visually “prettiest” to visually “most overcooked.” Subsequent eggs will be assigned numbers, which will be analyzed as a doe score. The higher the number, the poorer the yolk.<\/p>

To assess peeling quality, we will use time measurement. Using a stopwatch with a resolution of 0.01 s, we will measure:<\/p>

Y4 \u2013 Peeling Time [s].<\/strong> Theoretically, the shorter the better, although not necessarily. Therefore, in our doe, we will use a jewelry scale with a resolution of 0.01g to measure the next Y:<\/p>

Y5 \u2013 Shell mass [g]<\/strong>. Increased shell mass means that it did not separate “easily” from the egg white, and some of the egg white may have “gone with the shell.”<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\n
\n\t
\n\t\t\n\t\t

\n\t\t\t\t\t\t\t2. Other Ys, which may be affected by the tested factors. The idea is simple; it’s easy to improve one Y and ruin the ones that are currently good\t\n\t\t\t\t\n\t\t<\/h3>\n\t\t\n\t<\/div>\n<\/div>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Y6 \u2013 Weight loss after cooking [%]<\/strong>. Calculated as the ratio of the weight before to after cooking. Theoretically, the weight should not change because the egg only changes state, not volume. In practice, water vapor escapes through pores in the shell, and some of the egg white may dehydrate slightly. For an egg that does not crack (the desired situation), the weight decreases by approximately 0.5 to 2%. If it cracks, the loss will be uch greater. This Y does not directly assess the quality of the yolk or the ease of peeling the egg, but will be an indicator of the quality of the cooking process.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\n
\n\t
\n\t\t\n\t\t

\n\t\t\t\t\t\t\t3. Other measurements:\t\n\t\t\t\t\n\t\t<\/h3>\n\t\t\n\t<\/div>\n<\/div>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Depending on the situation, we may need to verify the correctness of certain test factors. This is necessary when the test factor has a certain inertia (for example, cooling and heating of a mold, e.g., a casting mold or an injection mold). Additionally, we may want to measure the level of certain interferences during the test, which may be useful during analysis.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\n
\n\t
\n\t\t\n\t\t

\n\t\t\t\t\t\t\tWhich experimental design should I choose? Full-factorial or fractional design?\t\n\t\t\t\t\n\t\t<\/h2>\n\t\t\n\t<\/div>\n<\/div>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

To test 6 factors at two levels, we have several options:<\/p>