Granted Patent B2

US 12,412,137 · App. 18/335,570 · Granted Sep 9, 2025

Inventors: Marcio Oliveira Almeida (Stittsville, CA); Zhen Lin (Kanata, CA); Behrouz Haji Soleimani (Halifax, CA); Seyednaser Nourashrafeddin (Ottawa, CA); Chantal Bisson-Krol (Kanata, CA)

Assignee: Kinaxis Inc.

G06Q10/06315G06N5/048G06N20/00G06F16/285

Abstract

Claims (54)

1. A computer-implemented method comprising:

receiving, by a processor, historical lead time data from a database;

dividing, by the processor, the historical lead time data into a first subset and a second subset;

training a plurality of machine learning forecasting algorithms on the first subset of the historical lead time data;

testing, each of the plurality of machine learning forecasting algorithms on the second subset of the historical lead time data and selecting, based on the testing, a machine learning forecasting algorithm having a highest accuracy;

retraining the selected machine learning forecasting algorithm based on the historical lead time data;

generating, by the selected machine learning forecasting algorithm, one or more lead time identifiers;

augmenting, by the processor, the historical lead time data with one or more features, the historical lead time data having a dimensionality greater than a 2-dimensional space and each of the one or more features having a dimensionality orthogonal to the dimensionality of the time historical lead time data;

projecting the historical lead time data to 2-dimensional space and reducing a noise thereof;

separating, by the processor, the historical lead time data into two or more groups based on a time density of data points comprising the historical lead time data;

selecting, by the processor, an optimal number of clusters for each of the two or more groups;

performing, by the processor, clustering on each of the two or more groups to provide a plurality of clusters, a number of clusters of the plurality of clusters for each of the two or more groups being the selected optimal number for that one of the two or more groups;

visually outputting the plurality of clusters;

receiving, by the processor, from a user via a user interface, tolerance criteria for separation of the historical lead time data into a plurality of tolerance zones;

separating, by the processor, the historical lead time data into the plurality of tolerance zones;

separating, by the processor, the plurality of clusters in accordance with a tolerance zone of each group; and

further separating, by the processor, the plurality of clusters according to the one or more lead time identifiers, to provide one or more separated clusters;

wherein separating the historical lead time data into two or more groups based on the time density comprises:

comparing a number of data points in each time series of the historical lead time data to at least one threshold, and

adding the data points of each of the time series into one group of the two or more groups based on the comparison to the at least one threshold.

2. The computer-implemented method of claim 1 , wherein the one or more features relate to linearity and seasonality.

3. The computer-implemented method of claim 2 , wherein the features are seasonality, upward linearity, flat linearity, and downward linearity.

4. The computer-implemented method of claim 1 , wherein the number of groups is three.

5. The computer-implemented method of claim 1 , further comprising:

pruning, by the processor, each cluster;

dividing, by the processor, each cluster into a plurality of subclusters.

6. A system comprising:

a processor; and

a memory storing instructions that, when executed by the processor, configure the system to:

receive, by the processor, historical lead time data from a database;

divide, by the processor, the historical lead time data into a first subset and a second subset;

train a plurality of machine learning forecasting algorithms on the first subset of the historical lead time data;

test, each of the plurality of machine learning forecasting algorithms on the second subset of the historical lead time data and select, based on the test, a machine learning forecasting algorithm having a highest accuracy;

retrain the selected machine learning forecasting algorithm based on the historical lead time data;

generate, by the selected machine learning forecasting algorithm, one or more lead time identifiers;

augment, by the processor, the historical lead time data with one or more features, the historical lead time data having a dimensionality greater than a 2-dimensional space and each of the one or more features having a dimensionality orthogonal to the dimensionality of the historical lead time data;

project the historical lead time data to 2-dimensional space and reduce a noise thereof;

separate, by the processor, the historical lead time data into two or more groups based on a time density of data points of the historical lead time data;

select, by the processor, an optimal number of clusters for each of the two or more groups;

perform, by the processor, clustering on each of the two or more groups to provide a plurality of clusters, a number of clusters of the plurality of clusters for each of the two or more groups being the selected optimal number for that one of the two or more groups;

visually output the plurality of clusters;

receive, by the processor, from a user via a user interface, tolerance criteria for separation of the historical lead time data into a plurality of tolerance zones;

separate, by the processor, the historical lead time data into the plurality of tolerance zones;

separate, by the processor, the plurality of clusters in accordance with a tolerance zone of each group; and

further separate, by the processor, the plurality of clusters according to the one or more lead time identifiers, to provide one or more separated clusters;

wherein the instructions that, when executed by the processor, configure the system to separate the historical lead time data into two or more groups based on the time density comprise instructions that, when executed by the system, configure the system to:

compare a number of data points in each time series of the historical lead time data to at least one threshold, and

add the data points of each of the time series into one group of the two or more groups based on the comparison to the at least one threshold.

7. The system of claim 6 , wherein the one or more features relate to linearity and seasonality.

8. The system of claim 7 , wherein the features are seasonality, upward linearity, flat linearity, and downward linearity.

9. The system of claim 6 , wherein the number of groups is three.

10. The system of claim 6 , wherein the instructions further configure the system to:

prune, by the processor, each cluster;

divide, by the processor, each cluster into a plurality of subclusters.

Assignments (1)

Continuity (5)

References Cited (95)