Optimized routing of interactions to contact center agents based on machine learning

Patent metadata

Field	Value
Publication number	US-9635181-B1
Application number	US-201514887297-A
Country	US
Kind code	B1
Filing date	Oct 19, 2015
Priority date	Oct 19, 2015
Publication date	Apr 25, 2017
Grant date	Apr 25, 2017

Abstract

A system that is adapted to route interactions to contact center agents. More specifically, the system is adapted to identify an interaction to be routed, and identify a group of agents based on one or more constraints for generating one or more candidate agents. The system is also adapted to gather context data surrounding the candidate agents. For each agent of the candidate agents, the system is adapted to estimate an expected value to be obtained by routing the interaction to the agent. The system is further adapted to select a particular agent of the candidate agents based on the estimates, and signal a routing device for routing the interaction to the particular agent.

First claim

The invention claimed is:

1. A system for routing interactions to contact center agents, the system comprising:
a processor; and
a memory, wherein the memory has stored therein instructions that, when executed by the processor, cause the processor to:
identify an interaction to be routed;
identify a group of agents based on one or more constraints for generating one or more candidate agents;
gather context data surrounding the candidate agents;
for each agent of the candidate agents, estimate an expected value to be obtained by routing the interaction to the agent;
select, based on the estimates, a particular agent of the candidate agents that does not have a highest expected value from among the candidate agents; and
signal a routing device for routing the interaction to the particular agent.

2. The system of claim 1 , wherein the one or more constraints are skills for handling the plurality of interactions.

3. The system of claim 1 , wherein the expected value is calculated based on profile of the candidate agents, profile of a customer associated with the interaction, and intent of the interaction.

4. The system of claim 3 , wherein the profile of each of the candidate agents includes a dynamically added skill, wherein the dynamically added skill is ignored or mapped to another skill, for calculating the expected value.

5. The system of claim 3 , wherein a proficiency level is associated with a skill, wherein the proficiency level is adjusted based on analysis of call transfers for each of the candidate agents.

6. The system of claim 3 , wherein the profile of the candidate agents includes preference of each of the agents in handling the interaction, wherein the instructions further cause the processor to:
generate a routing offer to one or more of the candidate agents based on the preference of the corresponding agents.

7. The system of claim 6 , wherein the candidate agents and customers associated with a plurality of interactions are modeled as a network of agents and customers, wherein a connection between a node representing one of the candidate agents and a node representing one of the customers is indicative of a fit between the one of the candidate agents and the one of the customers.

8. The system of claim 7 , wherein the instructions that cause the processor to select the particular agent further include instructions that cause the processor to determine how well a customer associated with the interaction to be routed fits with the particular agent.

9. The system of claim 8 , wherein the instructions that cause the processor to determine how well the customer fits with the particular agent further include instructions that cause the processor to predict sentiment to be expected during the interaction between the customer and the particular agent.

10. The system of claim 1 , wherein the expected value is modeled as a standard normal distribution with an upper confidence bound, and the instructions that cause the processor to select the particular agent cause the processor to select an agent associated with an expected value with a highest upper confidence bound.

11. The system of claim 10 , wherein the selection of the agent associated with the highest upper confidence bound balances exploration and exploitation needs.

12. The system of claim 11 , wherein the exploration needs are satisfied by selecting agents that are deemed sub-optimal, and exploitation needs are satisfied by selecting agents that maximize expected values.

13. The system of claim 1 , wherein the instructions that cause the processor to select the particular agent include instructions that cause the processor to:
concurrently identify a plurality of interactions waiting to be routed, wherein the plurality of interactions include the interaction to be routed; and
select contact center agents for the plurality of interactions.

14. The system of claim 13 further comprising:
estimate the expected value to be obtained for routing each of the plurality of interactions to each of the contact center agents.

15. The system of claim 14 further comprising:
calculate a predicted wait time associated with each of the contact center agents, wherein the expected value for a particular one of the contact center agents is a function of the predicted wait time for the particular one of the plurality of agents.

16. The system of claim 1 , wherein the interactions are real-time interactions.

Assignees

• Genesys Telecommunications Laboratories Inc

Inventors

• Mcgann Conor
• Aravamudhan Bharath
• Makagon Petr
• Ristock Herbert Willi Artur
• Konig Yochai
• Duclos Gregory
• Zhakov Vyacheslav
• Pelemis Damjan

Classifications

• H04M2203/40

  related to call centers · CPC title

• H04M3/5233 Primary

  Operator skill based call distribution · CPC title

• H04M2203/401

  Performance feedback · CPC title

• H04M3/5232

  Call distribution algorithms · CPC title

• H04M3/5238

  with waiting time or load prediction arrangements · CPC title