Automatically prioritizing supply chain-related demand using artificial intelligence techniques

Patent metadata

Field	Value
Publication number	US-12159257-B2
Application number	US-202217575808-A
Country	US
Kind code	B2
Filing date	Jan 14, 2022
Priority date	Jan 14, 2022
Publication date	Dec 3, 2024
Grant date	Dec 3, 2024

Abstract

Methods, apparatus, and processor-readable storage media for automatically prioritizing supply chain-related demand using artificial intelligence techniques are provided herein. An example computer-implemented method includes processing supply-chain related data using a first set of artificial intelligence techniques trained based at least in part on historical demand availability data; processing supply-chain related data using a second set of artificial intelligence techniques trained based at least in part on historical supply availability data; processing supply-chain related data using a third set of artificial intelligence techniques trained based at least in part on historical production availability data; prioritizing multiple orders within a supply chain environment by processing, using a fourth set of artificial intelligence techniques, results from the first set of artificial intelligence techniques, the second set of artificial intelligence techniques, and the third set of artificial intelligence techniques; and performing one or more automated actions based on the prioritization of the multiple orders.

First claim

What is claimed is:

1. A computer-implemented method comprising:
processing at least a portion of supply-chain related data using a first set of artificial intelligence techniques trained based at least in part on historical demand availability data related to at least one supply chain environment;
processing at least a portion of the supply-chain related data using a second set of artificial intelligence techniques trained based at least in part on historical supply availability data related to the at least one supply chain environment;
processing at least a portion of the supply-chain related data using a third set of artificial intelligence techniques trained based at least in part on historical production availability data related to the at least one supply chain environment;
prioritizing multiple orders within the at least one supply chain environment by processing, using a fourth set of artificial intelligence techniques, at least a portion of results from the first set of artificial intelligence techniques, at least a portion of results from the second set of artificial intelligence techniques, and at least a portion of results from the third set of artificial intelligence techniques, wherein the fourth set of artificial intelligence techniques comprises at least one neural network trained using historical results from the first set of artificial intelligence techniques, historical results from the second set of artificial intelligence techniques, and historical results from the third set of artificial intelligence techniques; and
performing one or more automated actions based at least in part on the prioritization of the multiple orders;
wherein the method is performed by at least one processing device comprising a processor coupled to a memory.

2. The computer-implemented method of claim 1 , wherein the at least one neural network comprises an input layer, one or more hidden layers, and an output layer, wherein the one or more hidden layers comprise at least one rectified linear unit activation function, and wherein the output layer comprises a sigmoid activation function as binary classification model trained to output an indication that a given input order is prioritized or an indication that the given input order is not prioritized.

3. The computer-implemented method of claim 1 , wherein processing, using the fourth set of artificial intelligence techniques, comprises attributing distinct weights to the at least a portion of results from the first set of artificial intelligence techniques, the at least a portion of results from the second set of artificial intelligence techniques, and the at least a portion of results from the third set of artificial intelligence techniques.

4. The computer-implemented method of claim 3 , wherein attributing the distinct weights is based at least in part on one or more of region-related information, season-related information, and trend-related information.

5. The computer-implemented method of claim 1 , wherein the results from the first set of artificial intelligence techniques comprises a first set of prioritization predictions pertaining to the multiple orders, the results from the second set of artificial intelligence techniques comprises a second set of prioritization predictions pertaining to the multiple orders, and the results from the third set of artificial intelligence techniques comprises a third set of prioritization predictions pertaining to the multiple orders, and wherein processing, using the fourth set of artificial intelligence techniques, comprises aggregating the first set of prioritization predictions, the second set of prioritization predictions, and the third set of prioritization predictions.

6. The computer-implemented method of claim 1 , wherein performing one or more automated actions comprises allocating resources within the at least one supply chain environment in response to demand based at least in part on the prioritization of the multiple orders.

7. The computer-implemented method of claim 1 , wherein performing one or more automated actions comprises training the fourth set of artificial intelligence techniques based at least in part on information derived from the prioritization of the multiple orders.

8. The computer-implemented method of claim 1 , wherein one or more of the first set of artificial intelligence techniques, the second set of artificial intelligence techniques, and the third set of artificial intelligence techniques comprise at least one machine learning-based classification algorithm.

9. The computer-implemented method of claim 1 , wherein processing, using the fourth set of artificial intelligence techniques, comprises using one or more of at least one gradient descent-based optimization algorithm, at least one loss function as an optimization algorithm, and using root mean squared propagation as an optimization algorithm.

10. A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device:
to process at least a portion of supply-chain related data using a first set of artificial intelligence techniques trained based at least in part on historical demand availability data related to at least one supply chain environment;
to process at least a portion of the supply-chain related data using a second set of artificial intelligence techniques trained based at least in part on historical supply availability data related to the at least one supply chain environment;
to process at least a portion of the supply-chain related data using a third set of artificial intelligence techniques trained based at least in part on historical production availability data related to the at least one supply chain environment;
to prioritize multiple orders within the at least one supply chain environment by processing, using a fourth set of artificial intelligence techniques, at least a portion of results from the first set of artificial intelligence techniques, at least a portion of results from the second set of artificial intelligence techniques, and at least a portion of results from the third set of artificial intelligence techniques, wherein the fourth set of artificial intelligence techniques comprises at least one neural network trained using historical results from the first set of artificial intelligence techniques, historical results from the second set of artificial intelligence techniques, and historical results from the third set of artificial intelligence techniques; and
to perform one or more automated actions based at least in part on the prioritization of the multiple orders.

11. The non-transitory processor-readable storage medium of claim 10 , wherein processing, using the fourth set of artificial intelligence techniques, comprises attributing distinct weights to the at least a portion of results from the first set of artificial intelligence techniques, the at least a portion of results from the second set of artificial intelligence techniques, and the at least a portion of results from the third set of artificial intelligence techniques.

12. The non-transitory processor-readable storage medium of claim 10 , wherein the results from the first set of artificial intelligence techniques comprises a first set of prioritization predictions pertaining to the multiple orders, the results from the second set of artificial intelligence techniques comprises a second set of prioritization predictions pertaining to the multiple orders, and the results from the third set of artificial intelligence techniques comprises a third set of pri

Assignees

• Dell Products Lp

Inventors

• Kumar Dhilip S
• Sahoo Sujit Kumar
• Mohanty Bijan Kumar
• Dinh Hung T

Classifications

• G06N3/04

  Architecture, e.g. interconnection topology · CPC title

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N3/08

  Learning methods · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06Q10/06315

  Needs-based resource requirements planning or analysis · CPC title