METHOD AND SYSTEM TO ASSOCIATE DEVICE MODELS AND ACCOUNTS BY UTILIZATION BANDS

• US 20110082722A1
• Filed: 10/07/2009
• Published: 04/07/2011
• Est. Priority Date: 10/07/2009
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Handley, John C., Earl, Jeffrey R.
• Granted Patent: US 8,234,138 B2
• Time in Patent Office: Days
• US Class Current: 705/7.38
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06Q 10/0631 Resource planning, allocation, distributing or scheduling for enterprises or organisations
• CPC Class Codes: G06Q 10/06312 Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
• CPC Class Codes: G06Q 10/0639 Performance analysis of employees; Performance analysis of enterprise or organisation operations
• CPC Class Codes: G06Q 10/0875 Itemisation or classification of parts, supplies or services, e.g. bill of materials
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 50/06 Energy or water supply

Abstract

Methods and systems for identifying devices having a specified utilization are disclosed. A selected minimum utilization and maximum utilization defining a utilization band may be received. Device information for each of a plurality of accounts, including a device model and utilization information for each device associated with an account, may also be received. A utilization may be determined for each device based on the utilization information. For each device model in an account, a number of devices in an account having the device model and a utilization in the utilization band may be determined. A plurality of clusters may be identified for device model-account pairs, and at least one device model-account pair may be assigned to a cluster.

Claims

1. A system for identifying devices having a utilization within a utilization band, the system comprising:

a processor; and

a processor-readable memory in operable communication with the processor,

wherein the processor-readable memory contains one or more programming instructions for performing the following;

receiving a selection for a utilization band comprising a minimum utilization and a maximum utilization,

receiving device information for each of a plurality of accounts, wherein the device information for each account comprises a device model and utilization information for each device associated with the account,

for each device, determining a utilization based on the utilization information,

for each device model in each account, determining a number of devices in the account having the device model and for which the utilization is in the utilization band, wherein each combination of device model and account comprises a device model-account pair,

identifying a plurality of clusters for the device model-account pairs, and

assigning at least one device model-account pair to a cluster.

2. The system of claim 1, wherein the one or more programming instructions for determining a utilization based on the utilization information comprises one or more programming instructions for performing the following:

determining utilization values for a device for a plurality of time periods; and

determining a utilization for the device based on the utilization values.

3. The system of claim 1, wherein the one or more programming instructions for identifying a plurality of clusters for the device model-account pairs comprise one or more programming instructions for performing the following:

for each of a plurality of numbers of clusters;

determining a probability that each device model-account pair occurs,

determining a maximum log likelihood value equal to the sum, over all device model-account pairs, of the product of the number of devices in a device model-account pair having the device model and for which the utilization is in the utilization band and the log of the probability that the device model-account pair occurs, and

determining a penalized likelihood value based on at least the number of clusters and the maximum log likelihood value for the number of clusters; and

selecting a number of clusters from the plurality of numbers of clusters corresponding to the smallest penalized likelihood value.

4. The system of claim 3, wherein the one or more programming instructions for determining a penalized likelihood value comprise one or more programming instructions for determining K log|X|−

2l_mle(K), where;

K is the number of clusters,

|X| is the sum of the number of devices in each device model-account pair having the device model and for which the utilization is in the utilization band, and

l_mle(K) is the maximum log likelihood value for K clusters.

5. The system of claim 3, wherein the one or more programming instructions for determining a penalized likelihood function value comprise one or more programming instructions for determining 2K−

2l_mle(K), where;

K is the number of clusters, and

l_mle(K) is the maximum log likelihood value for K clusters.

6. The system of claim 1, wherein the one or more programming instructions for assigning at least one device model-account pair to a cluster comprise one or more programming instructions for performing the following:

for at least one device model-account pair,

for each cluster, determining a probability that the device model-account pair is associated with the cluster, and

in response to at least one probability for the device model-account pair surpassing a threshold value, assigning the device model-account pair to the cluster associated with the highest probability.

7. The system of claim 1, wherein the one or more programming instructions for assigning at least one device model-account pair to a cluster comprise one or more programming instructions for performing the following:

for at least one device model-account pair,

for each cluster, determining a probability that the device model-account pair is associated with the cluster, and

assigning the device model-account pair to the cluster associated with the highest probability.

8. The system of claim 1, wherein the processor-readable storage medium further contains one or more programming instructions for determining a confidence value for the at least one device model-account pair assigned to a cluster.

9. The system of claim 1, wherein the processor-readable storage medium further contains one or more programming instructions for displaying the device model-account pairs associated with the cluster.

10. A method of identifying devices having a utilization within a utilization band, the method comprising:

receiving, by a processing device, a selection for a utilization band comprising a minimum utilization and a maximum utilization;

receiving, by the processing device, device information for each of a plurality of accounts, wherein the device information for each account comprises a device model and utilization information for each device associated with the account;

for each device model in each account, determining, via the processing device, a number of devices in the account having the device model and a utilization in the utilization band, wherein each combination of device model and account comprises a device model-account pair; and

identifying, via the processing device, a plurality of clusters for the device model-account pairs.

11. The method of claim 10, wherein determining a utilization based on the utilization information comprises:

determining utilization values for a device for a plurality of time periods; and

determining a utilization for the device based on the utilization values.

12. The method of claim 10, wherein identifying a plurality of clusters for the device model-account pairs comprises:

for each of a plurality of numbers of clusters;

determining a probability that each device model-account pair occurs,

determining a maximum log likelihood value equal to the sum, over all device model-account pairs, of the product of the number of devices in a device model-account pair having the device model and for which the utilization is in the utilization band and the log of the probability that the device model-account pair occurs, and

determining a penalized likelihood value based on at least the number of clusters, the maximum log likelihood value for the number of clusters; and

selecting a number of clusters from the plurality of numbers of clusters corresponding to the smallest penalized likelihood value.

13. The method of claim 12, wherein determining a penalized likelihood value comprises determining K log|X|−

2l_mle(K), where;

K is the number of clusters,

|X| is the sum of the number of devices in a device model-account pair having the device model and for which the utilization is in the utilization band, and

l_mle(K) is the maximum log likelihood value for K clusters.

14. The method of claim 12, wherein determining a penalized likelihood value comprises determining 2K−

2l_mle(K), where;

K is the number of clusters, and

l_mle(K) is the maximum log likelihood value for K clusters.

15. The method of claim 10, wherein assigning at least one device model-account pair to a cluster comprises:

for at least one device model-account pair,

for each cluster, determining a probability that the device model-account pair is associated with the cluster, and

in response to at least one probability for the device model-account pair surpassing a threshold value, assigning the device model-account pair to the cluster associated with the highest probability.

16. The method of claim 10, wherein assigning at least one device model-account pair to a cluster comprises:

for at least one device model-account pair,

for each cluster, determining a probability that the device model-account pair is associated with the cluster, and

assigning the device model-account pair to the cluster associated with the highest probability.

17. The method of claim 10, further comprising:

determining a confidence value for the at least one device model-account pair assigned to a cluster.

18. The method of claim 10, further comprising:

displaying the device model-account pairs associated with at least one of the plurality of clusters.

19. A system for identifying devices having a utilization within a utilization band, the system comprising:

a processor; and

a processor-readable memory in operable communication with the processor,

wherein the processor-readable memory contains one or more programming instructions for performing the following;

receiving a selection for a utilization band comprising a minimum utilization and a maximum utilization,

receiving device information for each of a plurality of accounts, wherein the device information for each account comprises a device model and a utilization for each device associated with the account,

for each device model in each account, determining a number of devices in the account having the device model and for which the utilization is in the utilization band, wherein each combination of device model and account comprises a device model-account pair,

identifying a plurality of clusters for the device model-account pairs, and

assigning at least one device model-account pair to a cluster.

20. The system of claim 19, wherein the one or more programming instructions for identifying a plurality of clusters for the device model-account pairs comprise one or more programming instructions for performing the following:

for each of a plurality of numbers of clusters;

determining a probability that each device model-account pair occurs,

determining a maximum log likelihood value equal to the sum, over all device model-account pairs, of the product of the number of devices in a device model-account pair having the device model and for which the utilization is in the utilization band and the log of the probability that the device model-account pair occurs, and

determining a penalized likelihood value based on at least the number of clusters and the maximum log likelihood value for the number of clusters; and

selecting a number of clusters from the plurality of numbers of clusters corresponding to the smallest penalized likelihood value.