Systems and methods for inventory management and optimization

Patent metadata

Field	Value
Publication number	US-12555069-B2
Application number	US-202117244817-A
Country	US
Kind code	B2
Filing date	Apr 29, 2021
Priority date	Nov 1, 2018
Publication date	Feb 17, 2026
Grant date	Feb 17, 2026

Abstract

The present disclosure provides systems and methods that may advantageously apply machine learning to accurately manage and predict inventory variables with future uncertainty. In an aspect, the present disclosure provides a system that can receive an inventory dataset comprising a plurality of inventory variables that indicate at least historical (i) inventory levels, (ii) inventory holding costs, (iii) supplier orders, or (iv) lead times over time. The plurality of inventory variables can be characterized by having one or more future uncertainty levels. The system can process the inventory dataset using a trained machine learning model to generate a prediction of the plurality inventory variables. The system can provide the processed inventory dataset to an optimization algorithm. The optimization algorithm can predict one or more inventory management parameters that result in a particular probability of achieving a target service level while minimizing a cost. The optimization algorithm can comprise constraint conditions.

First claim

What is claimed is:

1 . A system comprising one or more computers and one or more storage devices having instructions stored thereon that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
receiving, by one or more processors, an inventory dataset comprising a plurality of inventory variables that indicate at least historical values of inventory variables; processing, by the one or more processors, the inventory dataset using a trained machine learning model to predict statistical distributions of future values of the plurality of inventory variables, wherein the trained machine learning model is trained using historical inventory datasets, and wherein the predicted statistical distributions reflect uncertainties of the future values of the plurality of inventory variables; simulating effects of the uncertainties of the future values of the plurality of inventory variables; applying, by the one or more processors, an optimization algorithm to at least the predicted statistical distributions of future values of the plurality of inventory variables to predict one or more inventory management parameters that achieve a service level constraint, while minimizing a cost function or maximizing a reward function, wherein the optimization algorithm comprises one or more constraint conditions, wherein the inventory management parameters comprise a timing and quantity of orders to be placed with suppliers, and wherein the optimization algorithm leverages the simulating of the effects of the uncertainties of the future values of the plurality of inventory variables; and re-training, by the one or more processors, the trained machine learning model using updated data that includes additional data not used during prior training of the trained machine learning model in response to determining reduced performance for predictions over a planning horizon associated with the trained machine learning model, wherein the planning horizon dynamically changes based on the updated data.

2 . The system of claim 1 , wherein the operations further comprise:
applying, by the one or more processors, the optimization algorithm to a current inventory level and a quantity of current supplier orders to predict the one or more inventory management parameters.

3 . The system of claim 2 , wherein the operations further comprise:
generating, by the one or more processors, an inventory management recommendation based at least in part on the predicted future values of the plurality of inventory variables or the predicted one or more inventory management parameters.

4 . The system of claim 3 , wherein the inventory management recommendation comprises a recommendation to increase or decrease a future inventory level relative to the current inventory level.

5 . The system of claim 3 , wherein the inventory management recommendation is generated in real time, substantially real-time, or just-in-time.

6 . The system of claim 1 , wherein the one or more constraint conditions comprise a requirement for a target service level to satisfy a present, incoming, or expected demand requirement, and wherein the operations further comprise:
computing the present, incoming, or expected demand requirement using forecasted demand for finished products and a bill of materials (BOM) for the finished products.

7 . The system of claim 6 , wherein the BOM comprises a dynamic hierarchical graph.

8 . The system of claim 1 , wherein the inventory dataset is received from a plurality of disparate sources selected from the group consisting of smart devices, sensors, enterprise systems, extraprise, and Internet sources.

9 . The system of claim 1 , wherein the plurality of inventory variables further comprises one or more inventory movements selected from the group consisting of arrival movements, consumption movements, blocked movements, and inter-factory movements.

10 . The system of claim 1 , wherein the plurality of inventory variables further comprises one or more of a demand forecast, a material consumption, a transit time, and a shipping cost.

11 . The system of claim 1 , wherein the trained machine learning model is selected from the group consisting of a support vector machine (SVM), a naive Bayes classifier, a linear regression model, a quantile regression model, a logistic regression model, a random forest model, a neural network, a Gaussian process, and a Dirichlet processes.

12 . The system of claim 1 , wherein the statistical distribution is a parametric distribution comprising one or more statistical parameters selected from the group consisting of a median, a mean, a mode, a variance, a standard deviation, a quantile, a measure of central tendency, a measure of variance, a range, a minimum, a maximum, an interquartile range, and a percentile.

13 . The system of claim 1 , wherein the statistical distribution is a parametric distribution selected from the group consisting of a Gaussian distribution, a Gamma distribution, and a Poisson distribution.

14 . The system of claim 1 , wherein the optimization algorithm minimizes the cost function.

15 . The system of claim 1 , wherein one or more constraint conditions comprise a member selected from the group consisting of an inventory constraint, a service level constraint, an arrival window constraint, an order size constraint, and an ordering window constraint.

16 . The system of claim 1 , wherein the optimization algorithm comprises a stochastic constrained optimization algorithm.

17 . The system of claim 16 , wherein the stochastic constrained optimization algorithm comprises a mixed integer linear programming (MILP) problem.

18 . The system of claim 1 , wherein the optimization algorithm comprises a technique selected from the group consisting of a grid search, a random search, and a Bayesian optimization search.

19 . The system of claim 1 , wherein the inventory dataset comprises a plurality of inventory variables for a single-echelon inventory, and wherein the single-echelon inventory corresponds to a single node or facility.

20 . The system of claim 1 , wherein the optimization algorithm is configured to perform multi-echelon inventory optimization (MEIO), wherein the inventory dataset comprises a plurality of inventory variables for a multi-echelon inventory, and wherein the multi-echelon inventory corresponds to a supply chain network comprising a plurality of nodes or facilities.

21 . The system of claim 20 , wherein the optimization algorithm is configured to model the plurality of inventory variables using a probabilistic graphical model of the supply chain network.

22 . The system of claim 20 , wherein the optimization algorithm is configured to search for parameters of interest concerning the supply chain network using Bayesian optimization.

23 . The system of claim 1 , wherein the optimization algorithm maximizes a product output.

24 . The system of claim 1 , wherein the optimization algorithm maximizes a profit.

25 . The system of claim 1 , further comprising using one or more inventory management parameters to predict future supply and demand behavior and future service level from the inventory dataset, thereby improving inventory management.

26 . The system of claim 1 , wherein the historical values of inventory variables comprise inventory levels, inventory holding cos

Assignees

• C3 Ai Inc

Inventors

• Ohlsson Henrik
• Bellala Gowtham
• Khoshfetratpakazad Sina
• Banerjee Dibyajyoti
• Krishnan Nikhil

Classifications

• G06Q10/067

  Enterprise or organisation modelling · CPC title

• G06Q10/06315

  Needs-based resource requirements planning or analysis · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06Q10/087 Primary

  Inventory or stock management, e.g. order filling, procurement or balancing against orders · CPC title