The key methodology that makes all the disciplines work together to yield the
best overall formulation is called Desirability. Desirability is the latest
advance in optimization technology.
SynGenics has applied the concept of desirability for many years to factor
market requirements and engineering information in the optimization of compounds
and analysis of technologies. Now SmartD Technology insures that the desirability
function has a derivative over the entire range of each property variable
eliminating unnatural distortion of the solution hypersurface.
Ease of Use
For the novice, no setup is required. For the expert user, there is great
flexibility of function shape implemented by passing the curve through user
defined set points in property units. There are no hard-to-understand shape
factors. Shown here is a Target desirability function for a property variable Mod10%.
The shape points are automatically placed in the center of the range as on the
right side of the curve, and can be moved when the practiced user has reason
to reshape the curve, as shown on the left side.
Once ImproveIt finds a solution, you can use its 3-dimensional graphics to gain insight into
the relationships it searched to find the best formulation. Sometimes this can be the most
valuable part of the exercise, because you finally can see what the PC has been wrestling
with in the realm of numbers! ImproveIt incorporates the wisdom of the formulator in an
easy-to-use, step-by-step approach. Several items have default values so the new user
can achieve success before deep knowledge is acquired. Of course, the more experience you
already have, the more benefit you will derive from using this application. Windows-based,
ImproveIt is compatible with 95, 98, NT, 2000, and XP. All the standard features,
such as cut and paste, importing and exporting data, and linking to other documents
are available in ImproveIt.
ImproveIt® is the advanced tool for statistical analysis and compounding
and formulation optimization.
ImproveIt brings together the following disciplines:
- Statistical Design of Experiments
Statistical Design of Experiments is a powerful body of statistically based techniques
that allow a scientist or engineer to efficiently characterize a design space and to
establish the link between cause and effect for processes that have multiple input
factors and associated interactions.
- Response Surface Methodology
In statistics, response surface methodology (RSM) explores the
relationships between several explanatory variables and one or more response variables.
- Multiple Regression is
a statistical technique that predicts values of one variable on the basis of two or more other variables;
- Variance Partition Analysis;
- Optimization to address the formulation problem