Method and apparatus for optimized production of sheet-metal parts

Patent metadata

Field	Value
Publication number	US-11586996-B2
Application number	US-202217939986-A
Country	US
Kind code	B2
Filing date	Sep 8, 2022
Priority date	Mar 13, 2020
Publication date	Feb 21, 2023
Grant date	Feb 21, 2023

Abstract

A method for optimizing production of sheet-metal parts, the production comprising cutting out and singularizing the sheet-metal parts and bending the sheet-metal parts, wherein the method includes: (A) training a neural network, which is executed on a Monte Carlo tree search framework, by means of supervised learning and self-play with reinforcement learning; (B) recording constraints for the sheet-metal parts, the constraints comprising geometric data of the sheet-metal parts; (C) creating an optimized production schedule by way of the neural network; and (D) outputting the production schedule.

First claim

The invention claimed is:

1. A method for optimizing production of sheet-metal parts, the production comprising cutting out and singularizing the sheet-metal parts and bending the sheet-metal parts, the method comprising: A) training a neural network, which is executed on a Monte Carlo tree search framework, using supervised learning and self-play with reinforcement learning;
B) recording constraints for the sheet-metal parts, the constraints comprising geometric data of the sheet-metal parts;
C) creating an optimized production schedule by way of the neural network; and
D) outputting the production schedule.

2. The method as claimed in claim 1 , wherein the production of the sheet-metal parts further comprises:
deburring the sheet-metal parts;
joining the sheet-metal parts;
coating the sheet-metal parts; and
assembling the sheet-metal parts.

3. The method as claimed in claim 1 , wherein the method steps A) to D) are performed using an algorithm based on AlphaGo or AlphaGo Zero and the algorithm comprises the neural network.

4. The method as claimed in claim 1 , wherein the training in the method step A) is performed using heuristically ascertained solutions from optimized production schedules.

5. The method as claimed in claim 4 , wherein the optimized production schedules comprise earliest due date solutions.

6. The method as claimed in claim 1 , wherein the optimization comprises both minimization of scrap and optimization of production time.

7. The method as claimed in claim 6 , wherein the constraints in the method step B) additionally comprise production deadlines for the sheet-metal parts.

8. The method as claimed in claim 7 , wherein the constraints in the method step B) additionally comprise the values of the sheet-metal parts.

9. The method as claimed in claim 8 , further comprising assigning a scrap score to the scrap and meeting the production deadline is assigned a production deadline score, wherein the scrap score is based on the value of the sheet-metal parts, the optimization minimizing both the scrap score and the production deadline score.

10. The method as claimed in claim 1 , wherein the method steps B) to D) are performed on an event-triggered basis, the event being read in via an event interface.

11. The method as claimed in claim 10 , wherein the event is in the form of a request for further machining of a sheet-metal part, in the form of production machine capacity that is being released, in the form of a production machine failure and/or in the form of an urgent job.

12. The method as claimed in claim 10 , wherein the event is triggered and read in via the event interface by a production machine, an indoor localization system and/or a manufacturing execution system.

13. The method as claimed in claim 1 , the method further comprising a method step E), which comprises a user rating of the production schedule that is output in method step D) being read in and the neural network being trained further with the user rating.

14. An apparatus for performing the method as claimed in claim 1 , wherein the apparatus comprises a computer for storing and executing the neural network, a constraint interface for reading in the constraints and a production schedule interface for outputting the production schedule.

15. The apparatus as claimed in claim 14 , further comprising an event interface, a production machine, an indoor localization system and/or a manufacturing execution system, an event triggered by the production machine, the indoor localization system and/or the manufacturing execution system being able to be read in via the event interface.

Assignees

• Trumpf Werkzeugmaschinen Se Co Kg

Inventors

• Mieth Carina
• Ottnad Jens
• Rinciog Alexandru
• Struckmeier Frederick

Classifications

• G06Q50/04

  Manufacturing · CPC title

• B21D28/26

  in sheets or flat parts · CPC title

• Y02P90/30

  Computing systems specially adapted for manufacturing · CPC title

• G05B13/027

  using neural networks only · CPC title

• G06Q10/063 Primary

  Operations research, analysis or management · CPC title