Patent · US Active

Methods and apparatus for machine learning predictions and multi-objective optimization of manufacturing processes

US10061300B1 · kind B1 · utility

60Cited by
3References
21Claims
0Family size

Assignee

Inventors

Key dates

Filing dateSep 29, 2017
Grant dateAug 28, 2018
Priority date
Expiry dateSep 29, 2037

Classification

  • Technology area (CPC Y)Emerging Cross-Sectional Technologies
  • CPC primaryY02P90/02
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

The subject technology is related to methods and apparatus for discretization, manufacturability analysis, and optimization of manufacturing process based on computer assisted design models and machine learning. An apparatus determines from the digital model features of a physical object. Thereafter, the apparatus produces predictive values for manufacturing processes based on regression machine learning models. The apparatus generates a non-deterministic response including a non-empty set of attributes of manufacture processes of the physical object based on a multi-objective optimization model. The non-deterministic response complies or satisfies a selected multi-objective condition included in the multi-objective optimization model.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.