Back to results
US 11,454,956 B2Granted
Manufacturing support system and method
AssigneeOMRON Corporation
InventorsAkio Sakumiya, Hiroyuki Miyaura, Michie Uno
Life of this patent
Estimating the term…
Abstract
A manufacturing support system may be provided. The manufacturing support system may comprise: an obtaining unit (IO) configured to obtain object data of an object to be manufactured; an artificial intelligence, Al, engine (20) configured to receive the object data as an input and to determine a hardware configuration of a manufacturing system for manufacturing the object with reference to information relating to available hardware for the manufacturing system; and an output unit (60) configured to output the determined hardware configuration.
Claims15
A manufacturing support system comprising:
- an obtaining unit configured to obtain object data of an object to be manufactured;
- an artificial intelligence, AI, engine configured to receive the object data as an input and to determine:
- manufacturing process steps to be carried out by a manufacturing system for manufacturing the object;
- and
- a hardware configuration of the manufacturing system for manufacturing the object with reference to information relating to available hardware for the manufacturing system;
- and
- an output unit configured to output the determined manufacturing process steps and the determined hardware configuration,
- wherein the information relating to available hardware for the manufacturing system includes information indicating, for at least part of possible manufacturing process steps, at least one hardware element that is available and that is required to perform the manufacturing process step,
- wherein the AI engine comprises:
- a machine learning device that is configured to:
- receive the object data as an input;
- perform computation using the received object data;
- and
- output information indicating at least one set of manufacturing process steps for manufacturing the object based on the computation;
- and
- a hardware information processing unit that is configured to determine the hardware configuration of the manufacturing system by identifying, with reference to the information relating to available hardware, at least one hardware element required to perform each of one or more steps included in said at least one set of manufacturing process steps for manufacturing the object,
- wherein the machine learning device comprises a neural network configured to perform the computation using the received object data, the neural network comprising an output layer including output nodes corresponding to possible sets of manufacturing process steps for manufacturing the object to be manufactured, each of the output nodes being configured to output a value representing a likelihood that the corresponding set of manufacturing process steps is appropriate for manufacturing the object,
- wherein the neural network has been trained using training data including object data of one or more objects and information indicating sets of manufacturing process steps for manufacturing the one or more objects,
- wherein the machine learning device is further configured to, for outputting the information indicating at least one set of manufacturing process steps for manufacturing the object based on the computation:
- identify a specified number of output nodes which have the highest values of likelihood that the corresponding sets of manufacturing process steps are appropriate for manufacturing the object;
- and
- output information indicating the sets of manufacturing process steps corresponding to the specified number of identified output nodes.
19 more elements
A computer-implemented method for supporting manufacturing, the method comprising:
- obtaining object data of an object to be manufactured;
- receiving, by an artificial intelligence, AI, engine, the object data as an input;
- determining, by the AI engine, manufacturing process steps to be carried out by a manufacturing system for manufacturing the object;
- determining, by the AI engine, a hardware configuration of the manufacturing system for manufacturing the object with reference to information relating to available hardware for the manufacturing system;
- and
- outputting the determined manufacturing process steps and the determined hardware configuration,
- wherein the information relating to available hardware for the manufacturing system includes information indicating, for at least part of possible manufacturing process steps, at least one hardware element that is available and that is required to perform the manufacturing process step,
- wherein said determining of the manufacturing process steps is performed by a machine learning device comprised in the AI engine, said determining of the manufacturing process steps comprises:
- receiving the object data as an input;
- performing computation using the received object data;
- and
- outputting information indicating at least one set of manufacturing process steps for manufacturing the object based on the computation,
- wherein said determining of the hardware configuration of the manufacturing system is performed by identifying, with reference to the information relating to available hardware, at least one hardware element required to perform each of one or more steps included in the at least one set of manufacturing process steps for manufacturing the object,
- wherein the machine learning device comprises a neural network configured to perform the computation using the received object data, the neural network comprising an output layer including output nodes corresponding to possible sets of manufacturing process steps for manufacturing the object to be manufactured, each of the output nodes being configured to output a value representing a likelihood that the corresponding set of manufacturing process steps is appropriate for manufacturing the object,
- wherein the method further comprises:
- training the neural network using training data including object data of one or more objects and information indicating sets of manufacturing process steps for manufacturing the one or more objects, and
- wherein said outputting of the information indicating at least one set of manufacturing process steps for manufacturing the object based on the computation comprises:
- identifying a specified number of output nodes which have the highest values of likelihood that the corresponding sets of manufacturing process steps are appropriate for manufacturing the object;
- and
- outputting information indicating the sets of manufacturing process steps corresponding to the specified number of identified output nodes.
17 more elements
A computer program product comprising computer-readable instructions that, when loaded and run on a computer, cause the computer to perform a method comprising:
- obtaining object data of an object to be manufactured;
- receiving, by an artificial intelligence, AI, engine, the object data as an input;
- determining, by the AI engine, manufacturing process steps to be carried out by a manufacturing system for manufacturing the object;
- determining, by the AI engine, a hardware configuration of the manufacturing system for manufacturing the object with reference to information relating to available hardware for the manufacturing system;
- and
- outputting the determined manufacturing process steps and the determined hardware configuration,
- wherein the information relating to available hardware for the manufacturing system includes information indicating, for at least part of possible manufacturing process steps, at least one hardware element that is available and that is required to perform the manufacturing process step,
- wherein said determining of the manufacturing process steps is performed by a machine learning device comprised in the AI engine, said determining of the manufacturing process steps comprises:
- receiving the object data as an input;
- performing computation using the received object data;
- and
- outputting information indicating at least one set of manufacturing process steps for manufacturing the object based on the computation,
- wherein said determining of the hardware configuration of the manufacturing system is performed by identifying, with reference to the information relating to available hardware, at least one hardware element required to perform each of one or more steps included in the at least one set of manufacturing process steps for manufacturing the object,
- wherein the machine learning device comprises a neural network configured to perform the computation using the received object data, the neural network comprising an output layer including output nodes corresponding to possible sets of manufacturing process steps for manufacturing the object to be manufactured, each of the output nodes being configured to output a value representing a likelihood that the corresponding set of manufacturing process steps is appropriate for manufacturing the object,
- wherein the method further comprises:
- training the neural network using training data including object data of one or more objects and information indicating sets of manufacturing process steps for manufacturing the one or more objects, and
- wherein said outputting of the information indicating at least one set of manufacturing process steps for manufacturing the object based on the computation comprises:
- identifying a specified number of output nodes which have the highest values of likelihood that the corresponding sets of manufacturing process steps are appropriate for manufacturing the object;
- and
- outputting information indicating the sets of manufacturing process steps corresponding to the specified number of identified output nodes.
17 more elements
Dependencies are inferred from the claim text.