System and method for identification of replacement parts using artificial intelligence/machine learning
Systems, methods, and computer-readable media to identify replacement parts for an electronic asset using artificial intelligence/machine learning (AI/ML). In at least one embodiment, part data may be provided to an input of a trained AI/ML parts similarity model that has been trained to determine similarities between parts, including the parts used in a plurality of electronic assets of the organization. The trained AI/ML parts similarity model uses the part data to provide information relating to parts that are compatible with the part. In certain embodiments, the information relating to compatible parts includes an identifier for compatible parts and a similarity score indicating how similar each compatible part is to the part that is to be used to service the electronic asset.
1. A computer-implemented method comprising:
obtaining part data for a part that is to be used to service an electronic asset of an organization;
providing the part data to an input of a trained AI/ML parts similarity model that has been trained to determine similarities between parts, including parts used in a plurality of electronic assets of the organization, wherein
the trained AI/ML parts similarity model is trained using processed feature data extracted from a plurality of data sources having data relating to part numbers and corresponding part specifications for a plurality of parts, including parts used in the plurality of electronic assets,
the trained AI/ML parts similarity model uses the part data to provide information relating to parts that are compatible with the part, and
the information relating to compatible parts includes an identifier for compatible parts and a similarity score indicating how similar each compatible part is to the part that is to be used to service the electronic asset, the similarity score representing a degree of compatibility between the part and the each compatible part; and
providing the information relating to compatible parts to a user to facilitate service of the electronic asset; and wherein
the plurality of data sources includes one or more of:
telemetry data for parts used in the plurality of electronic assets;
purchase orders for parts used in the plurality of electronic assets;
invoices for parts used in the plurality of electronic assets;
inventory data for parts used in the plurality of electronic assets;
service records relating to the plurality of electronic assets;
service records for parts used in the plurality of electronic assets; and
a parts catalog for parts used in the plurality of electronic assets; and,
training of the trained AI/ML parts similarity model comprises:
extracting and selecting feature data from the plurality of data sources;
executing an unsupervised AI/ML categorical model using the feature data to generate clusters of data corresponding to similar parts;
executing a hash operation on the data corresponding to similar parts;
executing a quantization operation on the hashed data;
executing a mapping operation on the hashed data to map similar parts to correspondingly similar bins based on results of the quantization operation, wherein the mapping operation generates binned part data including the hashed data and corresponding bins; and
using the binned part data as an input for training the trained AI/ML parts similarity model.
2. The computer-implemented method of claim 1 , wherein
the trained AI/ML parts similarity model further provides information relating to availability of one or more of the compatible parts.
3. The computer-implemented method of claim 1 , wherein
the trained AI/ML parts similarity model is further configured to identify alternative parts that are compatible with the part, wherein the alternative parts include parts not currently used in any of the plurality of electronic assets.
4. The computer-implemented method of claim 1 , further comprising:
automatically generating an order to obtain one or more of the compatible parts from a vendor.
5. The computer-implemented method of claim 1 , further comprising:
using a trained AI/ML parts failure model to predict failure of a part used in an electronic asset; and
providing part data for the part predicted to fail to an input of the trained AI/ML parts similarity model, wherein the trained AI/ML parts similarity model uses the part data for the part predicted to fail to provide information relating to parts that are compatible with the part predicted to fail.
6. A system comprising:
one or more information handling systems, wherein the one or more information handling systems include:
a processor;
a data bus coupled to the processor; and
a non-transitory, computer-readable storage medium embodying computer program code, the non-transitory, computer-readable storage medium being coupled to the data bus;
wherein the computer program code included in one or more of the information handling systems is executable by the processor of the information handling system so that the information handling system, alone or in combination with other information handling systems, executes operations comprising:
obtaining part data for a part that is to be used to service an electronic asset of an organization;
providing the part data to an input of a trained AI/ML parts similarity model that has been trained to determine similarities between parts, including parts used in a plurality of electronic assets of the organization, wherein
the trained AI/ML parts similarity model is trained using processed feature data extracted from a plurality of data sources having data relating to part numbers and corresponding part specifications for a plurality of parts, including parts used in the plurality of electronic assets,
the trained AI/ML parts similarity model uses the part data to provide information relating to parts that are compatible with the part, and
the information relating to compatible parts includes an identifier for compatible parts and a similarity score indicating how similar each compatible part is to the part that is to be used to service the electronic asset, the similarity score representing a degree of compatibility between the part and the each compatible part; and
providing the information relating to compatible parts to a user to facilitate service of the electronic asset; and wherein
the plurality of data sources includes one or more of:
telemetry data for parts used in the plurality of electronic assets;
purchase orders for parts used in the plurality of electronic assets;
invoices for parts used in the plurality of electronic assets;
inventory data for parts used in the plurality of electronic assets;
service records relating to the plurality of electronic assets;
service records for parts used in the plurality of electronic assets; and
a parts catalog for parts used in the plurality of electronic assets; and,
training of the trained AI/ML parts similarity model comprises:
extracting and selecting feature data from the plurality of data sources;
executing an unsupervised AI/ML categorical model using the feature data to generate clusters of data corresponding to similar parts;
executing a hash operation on the data corresponding to similar parts;
executing a quantization operation on the hashed data;
executing a mapping operation on the hashed data to map similar parts to correspondingly similar bins based on results of the quantization operation, wherein the mapping operation generates binned part data including the hashed data and corresponding bins; and
using the binned part data as an input for training the trained AI/ML parts similarity model.
7. The system of claim 6 , wherein
the trained AI/ML parts similarity model further provides information relating to availability of one or more of the compatible parts.
8. The system of claim 6 , wherein
the trained AI/ML parts similarity model is further configured to identify alternative parts that are compatible with the part, wherein the alternative parts include parts not currently used in any of the plurality of electronic assets.
9. The system of claim 6 , wherein the operations further comprise:
automatically generating an order to obtain one or more of the compatible parts from a vendor.
10. The system of claim 6 , wherein the operations further comprise:
using a trained AI/ML parts failure model to predict failure of a part used in an electronic asset; and
providing part data for the part predicted to fail to an input of the trained AI/ML parts similarity model, wherein the trained AI/ML parts similarity model uses the part data for the part predicted to fail to provide information relating to parts that are compatible with the part predicted to fail.
11. A non-transitory, computer-readable storage medium embodying computer program code, the computer program code comprising computer executable instructions configured for:
obtaining part data for a part that is to be used to service an electronic asset of an organization;
providing the part data to an input of a trained AI/ML parts similarity model that has been trained to determine similarities between parts, including parts used in a plurality of electronic assets of the organization, wherein
the trained AI/ML parts similarity model is trained using processed feature data extracted from a plurality of data sources having data relating to part numbers and corresponding part specifications for a plurality of parts, including parts used in the plurality of electronic assets,
the trained AI/ML parts similarity model uses the part data to provide information relating to parts that are compatible with the part, and
the information relating to compatible parts includes an identifier for compatible parts and a similarity score indicating how similar each compatible part is to the part that is to be used to service the electronic asset, the similarity score representing a degree of compatibility between the part and the each compatible part; and
providing the information relating to compatible parts to a user to facilitate service of the electronic asset; and wherein
the plurality of data sources includes one or more of:
telemetry data for parts used in the plurality of electronic assets;
purchase orders for parts used in the plurality of electronic assets;
invoices for parts used in the plurality of electronic assets;
inventory data for parts used in the plurality of electronic assets;
service records relating to the plurality of electronic assets;
service records for parts used in the plurality of electronic assets; and
a parts catalog for parts used in the plurality of electronic assets; and,
training of the trained AI/ML parts similarity model comprises:
extracting and selecting feature data from the plurality of data sources;
executing an unsupervised AI/ML categorical model using the feature data to generate clusters of data corresponding to similar parts;
executing a hash operation on the data corresponding to similar parts;
executing a quantization operation on the hashed data;
executing a mapping operation on the hashed data to map similar parts to correspondingly similar bins based on results of the quantization operation, wherein the mapping operation generates binned part data including the hashed data and corresponding bins; and
using the binned part data as an input for training the trained AI/ML parts similarity model.
12. The non-transitory, computer-readable storage medium of claim 11 , wherein
the trained AI/ML parts similarity model further provides information relating to availability of one or more of the compatible parts.
13. The non-transitory, computer-readable storage medium of claim 11 , wherein
the trained AI/ML parts similarity model is further configured to identify alternative parts that are compatible with the part, wherein the alternative parts include parts not currently used in any of the plurality of electronic assets.
14. The non-transitory, computer-readable storage medium of claim 11 , wherein the instructions are further configured for:
using a trained AI/ML parts failure model to predict failure of a part used in an electronic asset; and
providing part data for the part predicted to fail to an input of the trained AI/ML parts similarity model, wherein the trained AI/ML parts similarity model uses the part data for the part predicted to fail to provide information relating to parts that are compatible with the part predicted to fail.