Polyethylene Blend Prediction via Molecular Weight Correlation

The patent applies preliminary action by establishing correlations between easily measurable structural properties (molecular weight distribution breadth and molecular weight) and target properties (ESCR, FNCT) before actual blending occurs. This allows prediction of long-term stability properties without requiring extended testing, enabling verification to be performed quickly and cost-effectively.

The patent replaces mechanical testing systems with a predictive correlation model. Instead of relying on extensive physical testing to verify properties of recycled blends, the invention uses mathematical correlations between structural properties (measurable via standard techniques) and target properties, providing reliable predictions without requiring batch-by-batch extensive testing.

The patent creates a predictive model that copies the behavior of actual testing. By establishing correlations between structural properties and target properties from initial testing data, the model can predict long-term properties without requiring repeated extended testing for each batch, thus maintaining accuracy while reducing time and cost.

The patent applies parameter changes by focusing on specific structural parameters (molecular weight distribution breadth and molecular weight) that are easily measurable and have strong correlations with target properties. By changing the approach from measuring all possible properties to measuring these key structural parameters, the process remains simple while achieving accurate predictions.