Define Recipe problem in experimental design method

Recipe is one of the most important factors leading to successful food products. A recipe usually includes several ingredients, which have different effects on specific food quality. To study these effects is the prerequisite for being able to choose the optimal recipes.
 
Many food products are manufactured by mixing two or more ingredients. In bread and cake formulations, for example, flour, sugar, baking powder, shortening, and water are used. In this case, one or more properties of the food product generally depend only on the proportions of the ingredients present in the mixture and not on the amount of the mixture. One ingredient (an independent variable) cannot vary without changing at least one of the other ingredients in the mixture, because all the ingredients will be part of a constant sum of 100%. In other words, the variables or the ratios of different ingredients in the recipe are dependent on each other. These phenomena do not meet the orthogonality requirement of a conventional factorial design. Therefore, to study and model the effects that different ingredient components in a mixture have on the food product properties of interest, the factorial experimental  design is no longer suitable unless it is modified. The effect of ingredient components  (mixture variables) on food quality (response) are modeled differently from those effects based on the usual factorial experimental methodology.

than unrestrained amounts. These proportions are measured by volume, by weight, or by mole fraction. These are nonnegative numbers,  and, if expressed as fractions of the mixture, they must add up to a unity, especially if the ingredients to be studied are the only ingredients comprising the mixture.