US-12626221-B1 - Smart micro fulfillment centers powered by machine learning based clusters

Abstract

A system and method are disclosed for determining inventory to stock at a supply chain site. The method includes determining a set of inventory clusters identifying inventory items that have been purchased together, ranking the set of inventory clusters based on a number of times each cluster of inventory items was purchased together, determining an amount of shelf space needed for each of the set of inventory clusters, ranking any inventory clusters that were purchased an equal number of times based on the amount of shelf space needed to generate a cluster ranking, and determining inventory to stock at the supply chain site based on the cluster ranking. The method further includes generating an inventory stocking plan based on the determined inventory to stock, and implementing the inventory stocking plan at the supply chain site based on the inventory stocking plan using one or more pieces of automated stocking machinery.

Inventors

• Harish Kumar SOMISETTY
• Vamshi Krishnam Raju Bollepally
• Sudheer Kumar Gattu

Assignees

• Blue Yonder Group, Inc.

Dates

Publication Date

20260512

Application Date

20231215

Claims (20)

1. 1 . A system for determining inventory to stock at a supply chain site, comprising: a computer, comprising a processor and a memory, the computer configured to: determine a set of inventory clusters identifying inventory items that have been purchased together; rank the set of inventory clusters based on a number of times each cluster of inventory items was purchased together; determine an amount of shelf space needed for each of the set of inventory clusters; rank any inventory clusters that were purchased an equal number of times based on the amount of shelf space needed to generate a cluster ranking; determine inventory to stock at the supply chain site based on the inventory cluster ranking; and produce products, by automated robotic production machinery, based, at least in part, on the determined inventory.
2. 2 . The system of claim 1 , wherein the computer is further configured to: generate an inventory stocking plan based on the determined inventory to stock; and implement the inventory stocking plan at the supply chain site based on the inventory stocking plan using one or more pieces of automated stocking machinery.
3. 3 . The system of claim 1 , wherein the computer is further configured to: refine the cluster ranking according to one or more smart attributes, the one or more smart attributes being selected from: trends in target customers for the supply chain site; events influencing demand for the inventory items; a search history of the inventory items; and sentiment analysis of publicly-available messages associated with the inventory items.
4. 4 . The system of claim 3 , wherein the computer is further configured to: perform the sentiment analysis using a natural language processing model, the natural language processing model being selected from: a support vector machine, a term frequency model, a term frequency inverse document frequency model, a bag-of-words model, a logistic regression model, a Naïve Bayes model, a decision tree, a hidden Markov model, a convolutional neural network, a recurrent neural network, an auto-encoder model or a natural language processing transformer.
5. 5 . The system of claim 2 , wherein the determined inventory of the inventory stocking plan comprises one or more items within top ranked clusters, one or more items from a top percentage of ranked clusters, or a percentage threshold of the top ranked clusters.
6. 6 . The system of claim 2 , wherein the computer is further configured to: prioritize or deprioritize one or more items for the inventory stocking plan based on: one or more transportation constraints of the one or more items, one or more storage constraints of the supply chain site, one or more price constraints of the one or more items, or one or more other constraints of the one or more items or the supply chain site.
7. 7 . The system of claim 1 , wherein the inventory clusters comprise one or more combinations of items that have previously been purchased together.
8. 8 . A computer-implemented method for determining inventory to stock at a supply chain site, comprising: determining, by a computer comprising a processor and a memory, a set of inventory clusters identifying inventory items that have been purchased together; ranking, by the computer, the set of inventory clusters based on a number of times each cluster of inventory items was purchased together; determining, by the computer, an amount of shelf space needed for each of the set of inventory clusters; ranking, by the computer, any inventory clusters that were purchased an equal number of times based on the amount of shelf space needed to generate a cluster ranking; determining, by the computer, inventory to stock at the supply chain site based on the inventory cluster ranking; and producing products, by automated robotic production machinery, based, at least in part, on the determined inventory.
9. 9 . The computer-implemented method of claim 8 , further comprising: generating, by the computer, an inventory stocking plan based on the determined inventory to stock; and implementing, by the computer, the inventory stocking plan at the supply chain site based on the inventory stocking plan using one or more pieces of automated stocking machinery.
10. 10 . The computer-implemented method of claim 8 , further comprising: refining, by the computer, the cluster ranking according to one or more smart attributes, the one or more smart attributes being selected from: trends in target customers for the supply chain site; events influencing demand for the inventory items; a search history of the inventory items; and sentiment analysis of publicly-available messages associated with the inventory items.
11. 11 . The computer-implemented method of claim 10 , further comprising: performing, by the computer, the sentiment analysis using a natural language processing model, the natural language processing model being selected from: a support vector machine, a term frequency model, a term frequency inverse document frequency model, a bag-of-words model, a logistic regression model, a Naïve Bayes model, a decision tree, a hidden Markov model, a convolutional neural network, a recurrent neural network, an auto-encoder model or a natural language processing transformer.
12. 12 . The computer-implemented method of claim 9 , wherein the determined inventory of the inventory stocking plan comprises one or more items within top ranked clusters, one or more items from a top percentage of ranked clusters, or a percentage threshold of the top ranked clusters.
13. 13 . The computer-implemented method of claim 9 , further comprising: prioritizing or deprioritizing, by the computer, one or more items for the inventory stocking plan based on: one or more transportation constraints of the one or more items, one or more storage constraints of the supply chain site, one or more price constraints of the one or more items, or one or more other constraints of the one or more items or the supply chain site.
14. 14 . The computer-implemented method of claim 8 , wherein the inventory clusters comprise one or more combinations of items that have previously been purchased together.
15. 15 . A non-transitory computer-readable medium embodied with software for determining inventory to stock at a supply chain site, the software when executed is configured to: determine, by a computer comprising a processor and a memory, a set of inventory clusters identifying inventory items that have been purchased together; rank the set of inventory clusters based on a number of times each cluster of inventory items was purchased together; determine an amount of shelf space needed for each of the set of inventory clusters; rank any inventory clusters that were purchased an equal number of times based on the amount of shelf space needed to generate a cluster ranking; determine inventory to stock at the supply chain site based on the inventory cluster ranking; and produce products, by automated robotic production machinery, based, at least in part, on the determined inventory.
16. 16 . The non-transitory computer-readable medium of claim 15 , wherein the software when executed is further configured to: generate an inventory stocking plan based on the determined inventory to stock; and implement the inventory stocking plan at the supply chain site based on the inventory stocking plan using one or more pieces of automated stocking machinery.
17. 17 . The non-transitory computer-readable medium of claim 15 , wherein the software when executed is further configured to: refine the cluster ranking according to one or more smart attributes, the one or more smart attributes being selected from: trends in target customers for the supply chain site; events influencing demand for the inventory items; a search history of the inventory items; and sentiment analysis of publicly-available messages associated with the inventory items.
18. 18 . The non-transitory computer-readable medium of claim 17 , wherein the software when executed is further configured to: perform the sentiment analysis using a natural language processing model, the natural language processing model being selected from: a support vector machine, a term frequency model, a term frequency inverse document frequency model, a bag-of-words model, a logistic regression model, a Naïve Bayes model, a decision tree, a hidden Markov model, a convolutional neural network, a recurrent neural network, an auto-encoder model or a natural language processing transformer.
19. 19 . The non-transitory computer-readable medium of claim 16 , wherein the determined inventory of the inventory stocking plan comprises one or more items within top ranked clusters, one or more items from a top percentage of ranked clusters, or a percentage threshold of the top ranked clusters.
20. 20 . The non-transitory computer-readable medium of claim 16 , wherein the software when executed is further configured to: prioritize or deprioritize one or more items for the inventory stocking plan based on: one or more transportation constraints of the one or more items, one or more storage constraints of the supply chain site, one or more price constraints of the one or more items, or one or more other constraints of the one or more items or the supply chain site.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS The present disclosure is related to that disclosed in the U.S. Provisional Application No. 63/470,034, filed May 31, 2023, entitled “Smart Micro Fulfillment Centers Powered by Machine Learning Based Clusters.” U.S. Provisional Application No. 63/470,034 is assigned to the assignee of the present application. The present invention hereby claims priority under 35 U.S.C. § 119 (e) to U.S. Provisional Application No. 63/470,034. TECHNICAL FIELD The present disclosure relates generally to data processing and more specifically to processing data for determining inventory to stock at supply chain locations. BACKGROUND Many supply chains or other retail organizations utilize micro fulfillment centers (MFCs) as local hubs for inventory set up in accessible locations near populated customer bases. Supply chain operators set up MFCs as alternatives or supplements to traditional warehouses or distribution centers which are larger and typically in remote locations. Because MFCs typically have less available space than traditional warehouses, efficient usage of MFC space by stocking more profitable items that are more likely to be purchased by customers is important to supply chain operators. Existing methods for determining or recommending inventory for MFCs include manpower intensive and error-prone human planning and industry standard algorithms that only recommend individual items rather than item clusters. Market basket analysis can be used to recommend certain item clusters for MFCs, but fail to make second-order recommendations if the first-order ranking basis of number of orders is tied. Because existing MFC inventory planning is time consumer, error prone and limited in the types of inventory recommendations that can be made, existing MFC inventory planning systems are undesirable. BRIEF DESCRIPTION OF THE DRAWINGS A more complete understanding of the present invention may be derived by referring to the detailed description when considered in connection with the following illustrative figures. In the figures, like reference numbers refer to like elements or acts throughout the figures. FIG. 1 illustrates a supply chain network, in accordance with an embodiment; FIG. 2 illustrates smart MFC system 110, archiving system 120 and the planning and execution system of FIG. 1 in greater detail, in accordance with an embodiment; FIG. 3 illustrates an example method for determining what inventory to stock at a supply chain site based on item clusters, in accordance with an embodiment; and FIG. 4 illustrates a cluster diagram used to illustrate the ranking of clusters of purchased together items for a particular supply chain site, in accordance with an embodiment. DETAILED DESCRIPTION Aspects and applications of the invention presented herein are described below in the drawings and detailed description of the invention. Unless specifically noted, it is intended that the words and phrases in the specification and the claims be given their plain, ordinary and accustomed meaning to those of ordinary skill in the applicable arts. In the following description and for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the various aspects of the invention. It will be understood, however, by those skilled in the relevant arts, that the present invention may be practiced without these specific details. In other instances, known structures and devices are shown or discussed more generally in order to avoid obscuring the invention. In many cases, a description of the operation is sufficient to enable one to implement the various forms of the invention, particularly when the operation is to be implemented in software. It should be noted that there are many different and alternative configurations, devices and technologies to which the disclosed inventions may be applied. The full scope of the inventions is not limited to the examples that are described below. Disclosed herein are systems and methods for identifying clusters of products that are most likely to be ordered from a given micro fulfillment center (MFC), within a specified time frame. Embodiments may identify items with a highest degree of affinity with other items, to be present in the same order as a weighted sub-graph, based on machine learning (ML) or artificial intelligence (AI) techniques. Embodiments provide a flexible weighting mechanism that can utilize various item attributes to rank items clusters, such as the number of times an item is part of repeat orders, a catalogue definition of an item, price to profit ratios, transportation constraints, storage constraints and quantities ordered. Embodiments may identify a maximum number of item clusters that can be stored at a given MFC at a given time to be able to fulfill all items for as many orders as possible. Embodiments provide a completely automated process for identifying and ranking item clusters. Embodime