Method for improving prediction relating to the production of a polymeric product

A method for improving prediction relating to the production of a polymeric product, comprising providing a prediction model (1) for describing a functional relationship between production parameters (2), wherein the production parameters (2) comprise formulation portions data specifying raw material portions used for the production of a respective polymeric product and comprise processing parameters data specifying machine process properties during the production of that polymeric product, and product properties data (3) associated with that polymeric product on a computer system (5), which production parameters (2) and product properties data (3) form data entry properties for a respective polymeric product, providing user input comprising user product targets specifying a part of the data entry properties, wherein a new data entry (7) with data entry properties is generated by the computer system (5) for realizing the user product targets and entered into a formulation database of data entries with data entry properties and wherein for the new data entry (7) the specified data entry properties are determined, and generating the new data entry (7) by selecting a base data entry from the formulation database and modifying at least one data entry property of the base data entry based on applying the user product targets to the prediction model (1),

• wherein a reward metric (8) is determined by the computer system (5) based on the determined data entry properties,
• wherein based upon the reward metric (8) the prediction model (1) is updated by the computer system (5),
• wherein the user input comprises a user selection of raw materials from a list of raw materials predefined in the computer system (5), thereby defining combinations of the raw materials for a polymeric formulation,
• wherein for the new data entry (7) the formulation portions data specifies raw material portions from the user selection of raw materials,
• wherein the reward metric (8) is a proximity metric based on compliance between the determined data entry properties and the user product targets, and
• wherein the prediction model (1) is updated such that for a proximity metric indicating higher compliance, the modification of the modified at least one data entry property is reinforced and
• wherein for a proximity metric indicating a lower compliance, the modification of the at least one data property is weakened.