Consider the situation where a biscuit maker is looking to imitate a specific product from its main competitor. In many cases the ingredients could be up to 20, add to this the biscuit making process variables like dough mixing, molding, and oven temperature. It will be a Herculean task if not impossible for the R &D department to deliver the needed outcome without utilizing the Advanced Industrial Analytics tools.

The same applies to a flour mill, consider a situation where the technical department wants to know in advance how the change of the milling settings could affect the flour quality and what kind of additives could be used in order to get the specified rheological behavior.

MVA Methods like PCA are employed to describe the current state. These Data Mining tools allow to cope with multiple variables simultaneously and to ncover the influential variables and the relationship have with each other. This kind of analysis includes also methods like Classification, e.g. Cluster analysis, Regression analysis and Predictive Modeling. The Quality department has a better understanding of the product, the process, the relationship between the variables and the technical differences from the competition. Thus, enable them to define the areas of improvement.

Experimental Design Tactics like DOE are employed in order to reduce experimentation to the minimum and to uncover the important interactions in the system. This begins by investigating the various factors affecting the final result, such as process settings, raw materials blends, additives, and much more. The scope is to rate them according to their importance and to define the ‘sweet spot’. For example, to find which formula of additives will give the desired result for specified wheat quality and mill settings. These actions lead to the development of an operational envelope where the process and results are stable even when subjected to changes.