Quantisweb Optimization Software vs Statistical Software

M’Hammed Mountassir, PhD., Quantisweb Technologies Inc.

Statistical software such as Minitab, SAS, Splus, R or others are arsenals of all known statistical methods (parametric or nonparametric) that are either variate or multivariate that statisticians use either to model variables according to other variables (Modeling) or making inference to a population on the basis of the information obtained on a sample (Hypothesis tests). The results stemming from the use of software depend to a large extent on the expertise of the statistician and can answer only the questions that this statistician raises. All these statistical methods only deal with one property (Y) at a time.

The Quantisweb approach is an integrated approach based on 3 main mathematical methodologies Reference (1) that is not only multivariate (X) but also multi-targets (Y).

The first methodology is the use of a multi-criteria method to order the characteristics of the product according to their importance. This method is based on the principles of hierarchical process analysis (AHP) (Saaty cf reference (2)).

The second methodology is the use of an arsenal of statistical methods which allow the modeling of each of the characteristics of the product by contenting itself with a minimal DOE (non-factorial). These models are of all possible types (polynomial, logistic, …) linear or non-linear and this according to the information extracted automatically from the results of the experiments (ROE).

The third methodology is an optimization step, where the system receives information from both the user (the ideal values ​​of the properties or the deterministic physical laws) and the information of the two previous modules in order to establish a multifactorial and multidimensional function to optimize. The system then uses a hybrid optimization library which allows to optimize both random laws and / or deterministic laws simultaneously with or without constraints on the controllable variables, these constraints can be linear or conditional. This step also generates an optimal combination of values for ingredients and process variables that can be validated.

Quantisweb delivers its full potential when used in its entirety and allows the user to work in an optimal context generally unsuspected by the user (especially in the multidimensional case) and not be limited to the laws of behavior and then try to fit them to the outputs of a statistical software.

Minitab uses only the second methodology, but with a factorial DOE and for only one property at a time.

Reference (1): Summary of the Quantisweb patent. Internal document.
Reference (2): The Analytic Hierarchy Process, New York, McGraw Hill. Saaty, T.L (1980).