Agent rating prediction and routing

• US 9,860,380 B2
• Filed: 02/05/2014
• Issued: 01/02/2018
• Est. Priority Date: 01/29/2014
• Status: Active Grant

• Current Assignee: Avaya Incorporated
• Original Assignee: Avaya Incorporated
• Inventors: Shaffer, Shmuel, Tendick, Patrick, Davis, Sheldon
• Primary Examiner(s): AL AUBAIDI, RASHA S
• Application Number:
  US14/173,610
• Publication Number:
  US 20150215463A1
• Time in Patent Office: 1,427 Days
• Field of Search: 37926501-26502, 3792651, 37926512-26513
• CPC Class Codes: H04M 2203/401 Performance feedback
• CPC Class Codes: H04M 3/5232 Call distribution algorithms
• CPC Class Codes: H04M 3/5238 with waiting time or load prediction arrangements

Abstract

An agent rating prediction and routing mechanism provided by a contact center communication system for work assignment optimization is described along with various methods and mechanisms for administering the same. The prediction system proposed herein analyzes past agent performance, agent attributes, contact attributes, and customer attributes to calculate an outcome value and to provide a performance prediction for use in work item routing to contact center resources.

Claims

1. A method of operating a contact center, comprising:

receiving, at a processor of a work assignment engine a contact center and via a communication network, a work item from a first customer communication device associated with a first customer;

executing, with the processor, a prediction module, the executing comprising;

automatically predicting, with the processor, a first agent'"'"'s skill, wherein the first agent'"'"'s skill is predicted with a prediction model comprising a category model capable of predicting performance based solely on one or more attributes of the work item;

determining, with the processor, an uncertainty for the prediction of the first agent'"'"'s skill;

automatically predicting, with the processor, a second agent'"'"'s skill, wherein the second agent'"'"'s skill is predicted with the prediction model;

automatically comparing, with the processor, the prediction for the first agent with the prediction for the second agent;

based on the comparison of predictions and the determined uncertainty, identifying with the processor, one of the first agent and second agent as a preferred agent for handling the work item;

automatically assigning, with the processor, the work item to an agent communication device associated with the preferred agent; and

routing the work item from the work assignment to the agent communication device by establishing a connection between the first customer communication device and agent communication device, the connection enabling communication between the first customer communication device and agent communication device,

wherein the determined uncertainty results in the one of the first agent and the second agent being identified as the preferred agent.

2. The method of claim 1, wherein the prediction model used to predict the first and second agent'"'"'s skill is specific to a service category to which the work item is assigned.

3. The method of claim 2, wherein the service category is one of sales, support, or billing.

4. The method of claim 1, wherein the prediction of the first agent'"'"'s skill is based on the first agent'"'"'s past performance history on past work items that have one or more attributes in common with the work item.

5. The method of claim 1, further comprising:

using the prediction of the first agent'"'"'s skill to calculate, with the processor, at least one of a first predicted cost and first predicted revenue that the first agent will generate while processing the work item;

using the prediction of the second agent'"'"'s skill to calculate, with the processor, at least one of a second predicted cost and a second predicted revenue that the second agent will generate while processing the work item; and

comparing, with the processor, the at least one of a first predicted cost and first predicted revenue with the at least one of a second predicted cost and second predicted revenue.

6. The method of claim 1, wherein the determining, with the processor, an uncertainty for the prediction of the first agent'"'"'s skill comprises assessing a penalty against the prediction of the first agent'"'"'s skill, and further wherein the penalty is based on an experience level of the first agent.

7. A method of operating a contact center, comprising:

receiving, at a processor of a work assignment engine in a contact center and via a communication network, a plurality of work items from one or more customer communication devices;

determining, with the processor, an attribute combination for each of the plurality of work items, wherein the attribute combination for each of the plurality of work items comprises one or more attributes associated with contact processing requirements of each of the plurality of work items;

executing, with the processor, a prediction module, the executing comprising;

predicting, with the processor, an agent'"'"'s skill at processing a first work item in the plurality of work items, wherein the agent'"'"'s skill at processing the first work item is predicted with a prediction model specifically created based on one or more attributes of the agent and one or more attributes of the attribute combination of the first work item;

predicting, with the processor, the agent'"'"'s skill at processing a second work item in the plurality of work items, wherein the agent'"'"'s skill at processing the second work item is predicted with a prediction model specifically created based on one or more attributes of the agent and one or more attributes of the attribute combination of the second work item;

determining, with the processor, an uncertainty for the prediction of the agent'"'"'s skill at processing one or more of the plurality of work items;

comparing, with the processor, the prediction for the first work item with the prediction for the second work item;

based on the comparison of the predictions and the determined uncertainty, identifying, with the processor, one of the first and second work items as a preferred work item for the agent to handle;

assigning, with the processor, the preferred work item to an agent communication device of the agent; and

routing the preferred work item from the work assignment engine to the agent communication device by establishing a connection between the agent communication device and a preferred customer communication device corresponding to the preferred work item, the connection enabling communication between the preferred customer communication device and the agent communication device,

wherein the determined uncertainty results in the one of the first and second work items being identified as the preferred work item.

8. The method of claim 7, wherein the preferred work item is an optimal work item for the agent.

9. The method of claim 7, wherein the prediction model used to predict the agent'"'"'s skill at processing the first work item and the second work item is specific to past performance history on past work items that have one or more attributes in common with the first work item and the second work item, respectively.

10. The method of claim 7, further comprising:

using, with the processor, the prediction of the agent'"'"'s skill at processing the first work item to calculate at least one of a first predicted cost and first predicted revenue that the agent will generate while processing the first work item;

using, with the processor, the prediction of the agent'"'"'s skill at processing the second work item to calculate at least one of a second predicted cost and a second predicted revenue that the agent will generate while processing the second work item; and

comparing, with the processor, the at least one of a first predicted cost and first predicted revenue with the at least one of a second predicted cost and second predicted revenue.

11. The method of claim 7, wherein the determining, with the processor, an uncertainty for the prediction of the agent'"'"'s skill at processing one or more of the plurality of work items comprises assessing a penalty against the prediction, and further wherein the penalty is calculated based on a standard deviation associated with the prediction.

12. The method of claim 7, wherein the agent belongs to a service category comprising one of sales, support, or billing.

13. A server used in a contact center, the server comprising:

a work assignment engine comprising a processor; and

computer memory coupled with the processor, the computer memory comprising instructions that are executed by the processor, the instructions, when executed by the processor, enabling the processor to;

receive a work item via a communication network from a first customer communication device associated with a first customer in a contact center;

execute a prediction module, the executing comprising;

automatically predicting a first agent'"'"'s skill, wherein the first agent'"'"'s skill is predicted with a prediction model comprising a category model capable of predicting performance based solely on one or more attributes of the work item;

determining an uncertainty for the prediction of the first agent'"'"'s skill;

automatically predicting a second agent'"'"'s skill, wherein the second agent'"'"'s skill is predicted with the prediction model;

automatically comparing the prediction for the first agent with the prediction for the second agent;

based on the comparison of predictions and the determined uncertainty, identify one of the first agent and the second agent as a preferred agent for handing the work item;

automatically assign the work item to an agent communication device associated with the preferred agent; and

route the work item from the work assignment engine to the agent communication device by establishing a connection between the first customer communication device and the agent communication device, the connection enabling communication between the first customer communication device and the agent communication device,

wherein the determined uncertainty results in the one of the first agent and the second agent being identified as the preferred agent.

14. The server of claim 13, wherein the instructions further enable the processor to:

use the prediction of the first agent'"'"'s skill to calculate at least one of a first predicted cost and first predicted revenue that the first agent will generate while processing the work item;

use the prediction of the second agent'"'"'s skill to calculate at least one of a second predicted cost and a second predicted revenue that the second agent will generate while processing the work item; and

compare the at least one of a first predicted cost and first predicted revenue with the at least one of a second predicted cost and second predicted revenue.

15. The server of claim 13, wherein the prediction model used to predict the first and second agent'"'"'s skill is specific to a service category to which the work item is assigned, wherein the service category is one of sales, support, or billing, and wherein the prediction of the first agent'"'"'s skill is based on the first agent'"'"'s past performance history on past work items that have one or more attributes in common with the work item.

16. A server used in a contact center, the server comprising:

a work assignment engine comprising a processor; and

computer memory coupled with the processor, the computer memory comprising instructions that are executed by the processor, the instructions, when executed by the processor, enabling the processor to;

receive a plurality of work items via a communication network from one or more customer communication devices associated with one or more customers in a contact center;

determine an attribute combination for each of the plurality of work items, wherein the attribute combination for each of the plurality of work items comprises one or more attributes associated with contact processing requirements of each of the plurality of work items;

execute, with the processor, a prediction module, the executing comprising;

predicting a contact center agent'"'"'s skill at processing a first work item in the plurality of work items, wherein the contact center agent'"'"'s skill at processing the first work item is predicted with a prediction model specifically created based on one or more attributes of the contact center agent and one or more attributes of the attribute combination of the first work item;

predicting the contact center agent'"'"'s skill at processing a second work item in the plurality of work items, wherein the contact center agent'"'"'s skill at processing the second work item is predicted with a prediction model specifically created based on one or more attributes of the contact center agent and one or more attributes of the attribute combination of the second work item;

determining an uncertainty associated with the prediction of the contact center agent'"'"'s skill at processing each of the first and second work items;

comparing the prediction for the first work item with the prediction for the second work item;

determining an uncertainty associated with the prediction of the contact center agent'"'"'s skill at processing each of the first and second work items;

comparing the prediction for the first work item with the prediction for the second work item;

based on the comparison of the predictions and the determined uncertainty, identify one of the first and second work items as a preferred work item for the contact center agent to handle;

assign the preferred work item to an agent communication device associated with the contact center agent; and

route preferred work item to the agent communication device by establishing a connection between the agent communication device and a preferred customer communication device corresponding to the preferred work item, the connection enabling communication between the preferred customer communication device and the agent communication device,

wherein the determined uncertainty result in the one of the first and second work items being identified as the preferred work item.

17. The server of claim 16, wherein the preferred work item is an optimal work item for the contact center agent.

18. The server of claim 16, wherein the attributes of the contact center agent comprise one or more of language proficiency, product knowledge, or transaction type.

19. The server of claim 16, wherein the comparison of the prediction for the first work item with the prediction for the second work item includes use of the determined uncertainty for each of the prediction for the first work item and the prediction for the second work item.

20. The server of claim 16, wherein the contact center agent belongs to a service category comprising one of sales, support, or billing.