Dataset creation method, learning model generation method, non-transitory computer readable recording medium, and dataset creation device

A non-transitory computer readable recording medium storing a computer program for creating a dataset for machine learning by allowing a computer to repeatedly execute processing of:

• (a) changing a first molding condition parameter set in a molding machine such that quality of a molded product is degraded;
• (b) acquiring physical quantity data indicating a state of a molded product produced by the molding machine or a state of the molding machine in which the first molding condition parameter for degrading the quality of the molded product and a second molding condition parameter are set,
• wherein the second molding condition parameter differs from the first molding condition parameter;
• (c) changing the second molding condition parameter set in the molding machine;
• (d) acquiring the physical quantity data indicating the state of the molded product produced by the molding machine in which the first molding condition parameter or the state of the molding machine and the second molding condition parameter after change are set;
• and
• (e) storing the second molding condition parameter before change, the physical quantity data obtained when setting the second molding condition parameter, the second molding condition parameter after change or a change amount in the second molding condition parameter, and the physical quantity data obtained when setting the second molding condition parameter after change or an evaluation value based on the physical quantity data in association with each other,
• wherein steps (b)-(e) are iterated until the physical quantity data acquired in step (d) demonstrates an improvement in the state of the molded product produced by the molding machine or the state of the molding machine.

A dataset creation device, comprising

• a processor executing processing of creating a dataset for machine learning of a learning model for adjusting a molding condition parameter set in a molding machine,
• wherein the processor executes processing of:
• (a) changing a first molding condition parameter set in a molding machine such that quality of a molded product is degraded;
• (b) acquiring physical quantity data indicating a state of a molded product produced by the molding machine or a state of the molding machine in which the first molding condition parameter for degrading the quality of the molded product and a second molding condition parameter are set,
• wherein the second molding condition parameter differs from the first molding condition parameter;
• (c) changing the second molding condition parameter set in the molding machine;
• (d) acquiring the physical quantity data indicating the state of the molded product produced by the molding machine or the state of the molding machine in which the first molding condition parameter and the second molding condition parameter after change are set;
• (e) storing the second molding condition parameter before change, the physical quantity data obtained when setting the second molding condition parameter, the second molding condition parameter after change or a change amount in the second molding condition parameter, and the physical quantity data obtained when setting the second molding condition parameter after change or an evaluation value based on the physical quantity data in association with each other in a storage unit;
• and
• (f) creating a dataset for machine learning by repeating the change of the first and second molding condition parameters and the acquisition of the physical quantity data,
• wherein steps (b)-(e) are iterated until the physical quantity data acquired in step (d) demonstrates an improvement in the state of the molded product produced by the molding machine or the state of the molding machine.