IP Library Granted Patent US 12,632,622
Granted Patent B2
US 12,632,622 · App. 17/824,300 · Granted May 19, 2026

Automatic generation of bill of process from digital twin using exploratory simulation

Inventors: Chengtao Wen (Redwood City, CA); Juan L. Aparicio Ojea (Moraga, CA); Ines Ugalde Diaz (Redwood City, CA); Gokul Narayanan Sathya Narayanan (Emeryville, CA); Eugen Solowjow (Berkeley, CA); Wei Xi Xia (Daly City, CA); Yash Shahapurkar (Berkeley, CA); Shashank Tamaskar (Mohali, IN); Heiko Claussen (Wayland, MA)
Assignee: Siemens Aktiengesellschaft
G06F30/27G06F2119/18
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,632,622
App. No.
17/824,300
Granted
May 19, 2026
Kind
B2
Abstract

A method for automatically generating a bill of process in a manufacturing system comprising: receiving design information representative of a product to be produced; iteratively performing simulations of the manufacturing system; identifying manufacturing actions based on the simulations; optimizing the identified manufacturing actions to efficiently produce the product to be produced; generating, by the manufacturing system, a bill of process for producing the product. Simulations may be performed using a digital twin of the product being produced and a digital twin of the environment. System actions are optimized using a reinforcement learning technique to automatically produce a bill of process based on the design information of the product and task specifications.

Claims (31)

1 . A method for automatically generating a bill of process in a manufacturing system comprising:

receiving design information representative of a product to be produced;

receiving a digital twin relating to the product and a digital twin of an environment in which the manufacturing system operates;

using the digital twins for iteratively performing simulations of the manufacturing system;

receiving information relating to an uncertainty factor of the environment in which the manufacturing system operates, the uncertainty factor quantifying unpredictable time-varying conditions that affect a production process of the product to be produced;

evaluating manufacturing actions to determine successful solutions that achieve the design requirements according to the design information, the digital twin, and the information relating to the uncertainty factor of the environment;

identifying successful manufacturing actions based on the simulations;

optimizing the identified manufacturing actions to efficiently produce the product to be produced; and

generating, by the manufacturing system, a bill of process for producing the product, wherein the bill of process causes an industrial robot to execute commands for producing the product.

2 . The method according to claim 1 , wherein the manufacturing system comprises an autonomous machine for producing the product.

3 . The method according to claim 1 further comprising optimizing the identified manufacturing actions by processing received inputs in a reinforcement learning process.

4 . The method according to claim 1 wherein the bill of process comprises a list of components of the product in order of assembly and associated motion planning to assemble each component to make the product being produced.

5 . The method according to claim 1 , wherein the bill of process comprises a list of components in order of assembly.

6 . The method according to claim 3 , wherein the reinforcement learning process includes a neural network for establishing a policy defining a next action for an agent which acts to produce the product.

7 . The method according to claim 6 , further comprising training the neural network offline using simulation of the environment and the product.

8 . The method according to claim 6 , further comprising training the neural network during live production of the manufacturing system.

9 . The method according to claim 6 , further comprising:

labelling a candidate solution for manufacturing the product with a discrete classification of successful or unsuccessful as input to the neural network during training of the neural network.

10 . The method according to claim 6 , further comprising:

labelling a candidate solution for manufacturing the product as a continuous regression as inputs to the neural network during training of the neural network.

11 . The method according to claim 1 , further comprising:

optimizing the identified manufacturing actions based on one or more of the following factors:

forces due to friction;

forces attributable to gripping;

forces related to moving a component of the product relative to an axis; and

camera pose estimation errors.

12 . The method of claim 11 , wherein one or more of the factors of forces due to friction, forces attributable to gripping a component of the product, and forces related to moving a component of the product relative to an axis is ignored to speed up optimizing the identified manufacturing actions.

13 . The method according to claim 1 , wherein optimizing the identified manufacturing actions is based on applying a minimal amount of force on a component of the product and a shortening of cycle time used to position components of the product.

14 . The method according to claim 1 , wherein the bill of process comprises:

a list of components used to build the product in an order in which the components are assembled; and

a list containing a motion planning of each component as that component is assembled to make a product.

Assignments (4)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 24, 2022
From: SIEMENS CORPORATION
To: SIEMENS AKTIENGESELLSCHAFT
Reel/Frame 060881/0308 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 24, 2022
From: SIEMENS INDUSTRY, INC.
To: SIEMENS AKTIENGESELLSCHAFT
Reel/Frame 060881/0450 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 18, 2022
From: WEN, CHENGTAO; APARICIO OJEA, JUAN L.; DIAZ, INES UGALDE; SATHYA NARAYANAN, GOKUL NARAYANAN; SOLOWJOW, EUGEN; XIA, WEI XI; SHAHAPURKAR, YASH; TAMASKAR, SHASHANK
To: SIEMENS CORPORATION
Reel/Frame 060839/0740 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 15, 2022
From: CLAUSSEN, HEIKO
To: SIEMENS INDUSTRY, INC.
Reel/Frame 060804/0094 →
Priority Claims (1)
EP 21178050 · Jun 7, 2021 · regional
Continuity (1)
Related Publication 20220391565A1 · Dec 8, 2022
References Cited (8)
US 9811074B1 · Aichele et al. · 2017 [cited by applicant]
US 20210387350A1 · Oleynik · 2021 [cited by examiner]
CN 108628261A · 2018 [cited by applicant]
CN 109719730A · 2019 [cited by applicant]
CN 112784328A · 2021 [cited by applicant]
WO 2020167316A1 · 2020 [cited by applicant]
Naboni, R., Kunic, A., Kramberger, A., & Schlette, C. (Jan. 2, 2021). Design, simulation and robotic assembly of reversible timber structures. Construction Robotics, 5(1), 13-22. (Year: 2021). [cited by examiner]
European Search Report corresponding to application No. 21178050.7; 7 pages. [cited by applicant]