Determining component part combination using artificial intelligence
Described are techniques for artificial intelligence (AI) assisted recommendations for component parts for end-products that reduce the occurrence of a product failures. The techniques include obtaining measurement data for groups of component parts configured to be assembled as part of an end-product. The techniques further include obtaining specification scores for the component parts included in the groups of component parts based on the measurement data. The techniques further include selecting a component part combination from the groups of component parts using artificial intelligence analysis of the specification scores to determine that the component part combination decreases a probability of a product failure as compared to historical occurrences of the product failure. The techniques further include outputting information for at least one component part included in the component part combination.
1 . A computer-implemented method, comprising:
training an artificial intelligence model to learn failure patterns associated with component part combinations for an end-product, wherein the training configures the artificial intelligence model to determine component parts for a component part combination that reduce a probability of end-product failure attributable to the component part combination;
obtaining measurement data for groups of component parts configured to be assembled as part of the end-product, wherein the measurement data comprises measurement specifications and actual measurements for interfacing portions of component parts included in the groups of component parts;
obtaining specification scores for the component parts included in the groups of component parts, wherein the specification scores are based at least in part on the measurement specifications and the actual measurements for the interfacing portions of the component parts;
inputting the specification scores for the component parts to the artificial intelligence model to select a component part combination from the groups of component parts, wherein the artificial intelligence model evaluates the specification scores against the learned failure patterns to determine that the component part combination decreases a probability of a product failure as compared to historical occurrences of end-product failures attributable to other component part combinations; and
outputting information for at least one component part included in the component part combination to an automated system, wherein responsive to receiving the information, the automated system retrieves the at least one component part for assembly of the end-product.
2 . The computer-implemented method of claim 1 , wherein the component part combination forms a critical part group of the end-product that has been historically associated with the product failure.
3 . The computer-implemented method of claim 1 , wherein obtaining the measurement data for the groups of component parts further comprises:
identifying which of the groups of component parts are physically available for assembly of the end-product; and
obtaining the measurement data for the groups of component parts that are physically available for assembly of the end-product.
4 . The computer-implemented method of claim 1 , wherein obtaining the specification scores for the component parts further comprises:
assigning a specification score to each component part included in the groups of component parts to indicate where an actual measurement of an interfacing portion of the component part is located in a tolerance range defined in a specification for the component part.
5 . The computer-implemented method of claim 1 , wherein a training dataset for training the artificial intelligence model includes performance data and specification scores associated with the historical occurrences of the product failure.
6 . The computer-implemented method of claim 1 , further comprising receiving input for a first component part to include in the component part combination, wherein the artificial intelligence model performs a distance measurement to determine one or more additional component parts to include in the component part combination.
7 . The computer-implemented method of claim 6 , wherein the artificial intelligence model uses a Jaccard distance measurement to perform the distance measurement.
8 . The computer-implemented method of claim 1 , wherein outputting the information for the component part combination further comprises:
displaying the information in a user interface to allow physical retrieval of component parts that correspond to the information.
9 . A system, comprising:
one or more computer readable storage media storing program instructions; and
one or more processors which, in response to executing the program instructions, are configured to:
train an artificial intelligence model to learn failure patterns associated with component part combinations for an end-product, wherein the training configures the artificial intelligence model to determine component parts for a component part combination that reduce a probability of end-product failure attributable to the component part combination;
obtain measurement data for groups of component parts configured to be assembled as part of an end-product, wherein the measurement data comprises measurement specifications and actual measurements for interfacing portions of component parts included in the groups of component parts;
obtain specification scores for the component parts included in the groups of component parts, wherein the specification scores are based at least in part on the measurement specifications and the actual measurements for the interfacing portions of the component parts;
input the specification scores for the component parts to the artificial intelligence model to select a component part combination from the groups of component parts, wherein the artificial intelligence model evaluates the specification scores against the learned failure patterns to determine that the component part combination decreases a probability of a product failure as compared to historical occurrences of end-product failures attributable to other component part combinations; and
output information for at least one component part included in the component part combination to an automated system, wherein responsive to receiving the information, the automated system retrieves the at least one component part for assembly of the end-product.
10 . The system of claim 9 , wherein the component part combination forms a critical part group of the end-product that has been historically associated with the product failure.
11 . The system of claim 9 , wherein the instructions configured to cause the one or more processors to obtain the measurement data for the groups of component parts are further configured to:
query a parts catalog to determine which of the groups of component parts are physically available for assembly of the end-product; and
obtain the measurement data for the groups of component parts that are physically available for assembly of the end-product from the parts catalog.
12 . The system of claim 9 , wherein the instructions configured to cause the one or more processors to obtain the specification scores for the component parts are further configured to:
assign a specification score to each component part included in the groups of component parts to indicate where an actual measurement of an interfacing portion of the component part is located in a tolerance range defined in a specification for the component part.
13 . The system of claim 9 , wherein a training dataset for training the artificial intelligence model includes performance data and specification scores associated with the historical occurrences of the product failure.
14 . The system of claim 9 , further comprising receiving input for a first component part to include in the component part combination, wherein the artificial intelligence model performs a distance measurement to determine one or more additional component parts to include in the component part combination.
15 . The system of claim 9 , wherein the instructions configured to cause the one or more processors to output the information for the component part combination are further configured to:
send the information to an automated assembly system to enable the automated assembly system to retrieve component parts that correspond to the information and assemble the end-product using the component parts.
16 . A computer program product comprising one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions comprising instructions configured to cause one or more processors to:
train an artificial intelligence model to learn failure patterns associated with component part combinations for an end-product, wherein the training configures the artificial intelligence model to determine component parts for a component part combination that reduce a probability of end-product failure attributable to the component part combination;
obtain measurement data for groups of component parts configured to be assembled as part of an end-product, wherein the measurement data comprises measurement specifications and actual measurements for interfacing portions of component parts included in the groups of component parts;
obtain specification scores for the component parts included in the groups of component parts, wherein the specification scores are based at least in part on the measurement specifications and the actual measurements for the interfacing portions of the component parts;
input the specification scores for the component parts to the artificial intelligence model to select a component part combination from the groups of component parts, wherein the artificial intelligence model evaluates the specification scores against the learned failure patterns to determine that the component part combination decreases a probability of a product failure as compared to historical occurrences of end-product failures attributable to other component part combinations; and
output information for at least one component part included in the component part combination to an automated system, wherein responsive to receiving the information, the automated system retrieves the at least one component part for assembly of the end-product.
17 . The computer program product of claim 16 , wherein the instructions configured to cause the one or more processors to obtain the measurement data for the groups of component parts are further configured to:
receive the measurement data from a vendor of the groups of component parts; and
store the measurement data in a parts catalog that contains information for the groups of component parts that are physically available for assembly of the end-product.
18 . The computer program product of claim 16 , wherein the instructions are further configured to cause one or more processors to:
receive input for a first component part to include in the component part combination, wherein the artificial intelligence model performs a distance measurement to determine one or more additional component parts to include in the component part combination.
19 . The computer program product of claim 18 , wherein the input for the first component part is provided via a user interface for a replacement part service.
20 . The computer program product of claim 16 , wherein the specification scores for the component parts comprise a numerical value or a label indicating where an actual measurement for a component part is located in a tolerance range defined in a specification for the component part.