Decision management for items digitally represented in an information processing system

1 . A method comprising:

implementing a decision engine on a computing platform for executing a process for providing intelligent recommendations for managing an inventory of items by a given entity, wherein the process for providing intelligent recommendations for managing an inventory of items, comprises:

obtaining, for a set of items, data indicative of a previous percentage division between a first item ownership type and a second item ownership type for each item in the set of items, wherein the first item ownership type for a given item corresponds to items that are owned by the given entity, and the second item ownership type for the given item corresponds to items that are owned by another entity and potentially procurable by the given entity;

classifying each item in the set of items into one of a plurality of clusters, based on the obtained data, wherein each cluster represents a different percentage division range between the first item ownership type and the second item ownership type;

executing a first machine learning process that includes identifying any items in each cluster that deviate from the percentage division range for the cluster;

executing a second machine learning process for that includes applying weights to any identified items based on the deviation from the percentage division range;

re-classifying any identified items to another cluster in the plurality of clusters based on the applied weights;

training a third machine learning process based on the re-classified items; executing the trained third machine learning process that includes selecting and assigning a percentage division for a given item from within a percentage division range of a cluster in the plurality of clusters to which the given item corresponds based on one of a set of modes, wherein the set of modes comprises:

a first mode wherein the assigned percentage division is selected from a first portion of the percentage division range;

a second mode wherein the assigned percentage division is selected from a second portion of the percentage division range; and

a third mode wherein the assigned percentage division is selected from a third portion of the percentage division range; and

virtualizing the selected and assigned percentage division for the given item.

2 . The method of claim 1 , wherein the first mode comprises an aggressive mode, the second mode comprises a normal mode, and the third mode comprises a passive mode.

3 . The method of claim 1 , wherein executing the first machine learning process comprises executing an error classification algorithm.

4 . The method of claim 3 , wherein executing the error classification algorithm comprises executing a k-nearest neighbor algorithm.

5 . The method of claim 1 , wherein, for the given item, a deviation from the percentage division range for the cluster is based on an occurrence of an error condition defined by a set of error conditions.

6 . The method of claim 1 , wherein executing the second machine learning process comprises executing a weighted distance algorithm.

7 . The method of claim 1 , wherein:

the percentage division for the given item comprises a first percentage of a quantity of the given item having the first item ownership type and a second percentage of a quantity of the given item having the second item ownership type; and

a percentage division range represents a range for each of the first percentage and the second percentage.

8 . The method of claim 7 , wherein a deviation from a percentage division range comprises an error condition which indicates that more of the quantity of the given item having the first item ownership type was needed or that less of the quantity of the given item having the first item ownership type was needed.

9 . The method of claim 1 , wherein;

the given entity is an original equipment manufacturer;

the other entity is a vendor;

the given item represents a part that is used in a manufacturing process such that first item ownership type comprises ownership of a quantity of the part by the original equipment manufacturer that controls manufacturing of equipment using the part; and

the second item ownership type comprises ownership of the quantity of the part by the vendor of the part.

10 . An apparatus comprising:

at least one computing platform comprising at least one processor device coupled to at least one memory that stores computer program instructions which are executed by the at least one processor device to implement a decision engine that executes on the at least one computing platform to perform a process for providing intelligent recommendations for managing an inventory of items by a given entity, wherein the decision engine operates to:

obtain, for a set of items, data indicative of a previous percentage division between a first item ownership type and a second item ownership type for each item in the set of items, wherein the first item ownership type for a given item corresponds to items that are owned by the given entity, and the second item ownership type for the given item corresponds to items that are owned by another entity and potentially procurable by the given entity;

classify each item in the set of items into one of a plurality of clusters, based on the obtained data, wherein each cluster represents a different percentage division range between the first item ownership type and the second item ownership type;

execute a first machine learning process to identify any items in each cluster that deviate from the percentage division range for the cluster;

execute a second machine learning process to apply weights to any identified items based on the deviation from the percentage division range;

re-classify any identified items to another cluster in the plurality of clusters based on the applied weights

train a third machine learning process based on the re-classified items;

execute the trained third machine learning process to select and assign a percentage division for a given item from within a percentage division range of a cluster in the plurality of clusters to which the given item corresponds based on one of a set of modes, wherein the set of modes comprises:

a first mode wherein the assigned percentage division is selected from a first portion of the percentage division range;

a second mode wherein the assigned percentage division is selected from a second portion of the percentage division range; and

a third mode wherein the assigned percentage division is selected from a third portion of the percentage division range; and

virtualize the selected and assigned percentage division for the given item.

11 . The apparatus of claim 10 , wherein:

the percentage division for the given item comprises a first percentage of a quantity of the given item having the first item ownership type and a second percentage of a quantity of the given item having the second item ownership type; and

a percentage division range represents a range for each of the first percentage and the second percentage.

12 . A computer program product comprising 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:

implement a decision engine that executes on a computing platform to perform a process for providing intelligent recommendations for managing an inventory of items by a given entity, wherein the decision engine operates to:

obtain, for a set of items, data indicative of a previous percentage division between a first item ownership type and a second item ownership type for each item in the set of items, wherein the first item ownership type for a given item corresponds to items that are owned by the given entity, and the second item ownership type for the given item corresponds to items that are owned by another entity and potentially procurable by the given entity;

classify each item in the set of items into one of a plurality of clusters, based on the obtained data, wherein each cluster represents a different percentage division range between the first item ownership type and the second item ownership type;

execute a first machine learning process to identify any items in each cluster that deviate from the percentage division range for the cluster;

execute a second machine learning process to apply weights to any identified items based on the deviation from the percentage division range;

re-classify any identified items to another cluster in the plurality of clusters based on the applied weights

train a third machine learning process based on the re-classified items;

executing the trained third machine learning process to select and assign a percentage division for a given item from within a percentage division range of a cluster in the plurality of clusters to which the given item corresponds based on one of a set of modes, wherein the set of modes comprises:

a first mode wherein the assigned percentage division is selected from a first portion of the percentage division range;

a second mode wherein the assigned percentage division is selected from a second portion of the percentage division range; and

a third mode wherein the assigned percentage division is selected from a third portion of the percentage division range; and

virtualize the selected and assigned percentage division for the given item.

13 . The apparatus of claim 10 , wherein the first mode comprises an aggressive mode, the second mode comprises a normal mode, and the third mode comprises a passive mode.

14 . The apparatus of claim 10 , wherein the first machine learning process comprises an error classification algorithm.

15 . The apparatus of claim 14 , wherein the error classification algorithm comprises a k-nearest neighbor algorithm.

16 . The apparatus of claim 10 , wherein, for the given item, a deviation from the percentage division range for the cluster is based on an occurrence of an error condition defined by a set of error conditions.

17 . The apparatus of claim 10 , wherein the second machine learning process comprises a weighted distance algorithm.

18 . The apparatus of claim 11 , wherein a deviation from a percentage division range comprises an error condition which indicates that more of the quantity of the given item having the first item ownership type was needed or that less of the quantity of the given item having the first item ownership type was needed.

19 . The computer program product of claim 12 , wherein the first mode comprises an aggressive mode, the second mode comprises a normal mode, and the third mode comprises a passive mode.

20 . The computer program product of claim 12 , wherein:

the percentage division for the given item comprises a first percentage of a quantity of the given item having the first item ownership type and a second percentage of a quantity of the given item having the second item ownership type;

a percentage division range represents a range for each of the first percentage and the second percentage; and

a deviation from a percentage division range comprises an error condition which indicates that more of the quantity of the given item having the first item ownership type was needed or that less of the quantity of the given item having the first item ownership type was needed.