Method and system for dynamic adaptive routing of deferrable work in a contact center

1. A computer-implemented method for optimizing a workflow in a contact center in which deferrable work interactions are prioritized and assigned to agents for handling, wherein the method comprises the steps of:

providing:

a plurality of natural language processing (NLP) models, each NLP model being configured to accept text from a given deferrable work interaction as an input and generate NLP scores indicating how the given deferrable work interaction rates in accordance with a characteristic, wherein to generate the NLP scores of the given deferrable work interaction, each NLP model generates a sparse vector representation of the given deferrable work interaction;

a priority model configured to accept as inputs the scores generated from the plurality of NLP models and generate a priority score related to a priority characteristic indicating how the given deferrable work interaction should be prioritized for handling relative to other ones of the deferrable work interactions;

receiving the deferrable work interactions;

using text derived from the deferrable working interactions as inputs to the plurality of NLP models to generate the NLP scores for each of the deferrable work interactions;

using the generated NLP scores as inputs to the priority model to generate the priority score for each of the deferrable work interactions;

using the generated NLP scores to identify one or more candidate agents of the agents for handling each of the deferrable work interactions;

receiving an inbound work forecast for the contact center that predicts inbound work levels expected over one or more future work periods;

receiving, in relation to the one or more future work periods, agent work schedule data describing anticipated work schedules for the agents;

using an optimization process to generate an optimized workflow for the deferrable work interactions, wherein, for each of the deferrable work interactions, the optimized workflow comprises an assignment in which a selected agent is selected from the candidate agents for handling the deferrable work interaction and a target timeframe is scheduled for handling of the deferrable work interaction; and

routing each of the deferrable work interactions in accordance with the assignments of the optimized workflow;

wherein the optimization process is configured to optimize according to the following factors:

the priority score generated for each of the deferrable work interactions;

an expected availability over the one or more future work periods of the one or more candidate agents identified for each of the deferrable work interactions as determined from:

the agent work schedule data over the one or more future work periods; and

the predicted inbound work levels over the one or more future work periods given the inbound work forecast.

2. The method of claim 1 , wherein the deferrable work interactions each comprises an email communication sent by a customer that requests a response from the contact center.

3. The method of claim 2 , further comprising the step of:

determining, for each of the deferrable work interactions, a response deadline indicating a threshold for responding to the deferrable work interaction;

wherein the factors that the optimization process considers further includes the response deadline determined for each of the deferrable work interactions.

4. The method of claim 3 , further comprising the step of:

determining, for each of the deferrable work interactions, a negative economic impact to the contact center for not responding to the deferrable work interaction before the threshold;

wherein the factors that the optimization process considers further includes the negative economic impact determined for each of the deferrable work interactions.

5. The method of claim 4 , wherein the response deadline and the negative economic impact are determined according to terms defined in a service level agreement (SLA) that governs requirements as to how the contact center responds to the deferrable work interactions.

6. The method of claim 2 , wherein the priority model comprises:

an urgency component that provides an urgency score; and

an importance component that provides an importance score;

wherein the priority score is derived from a weighted combination of both the urgency score and the importance score.

7. The method of claim 2 , wherein the inbound work forecast comprises predictions as to an inbound work levels of immediate work interactions expected over the one or more future work periods;

wherein an immediate work interaction is defined as a synchronous interaction involving a real-time exchange of text or voice between one of the agents and a customer.

8. The method of claim 2 , wherein the step of using the generated NLP scores to identify the one or more candidate agents for handling each of the deferrable work interactions includes:

identifying preferred agent characteristics for each of the deferrable work interactions; and

comparing the identified preferred agent characteristics against actual agent characteristics of the agents and determining the one or more candidate agents as being ones of the agents having a most favorable match;

wherein the preferred agent characteristics are defined as being agent characteristics found to more likely produce a favorable outcome given a characteristic of a given deferrable work interaction.

9. The method of claim 2 , wherein the factors that the optimization process considers further includes:

a current backlog level of the deferrable work interactions.

10. The method of claim 2 , wherein the plurality of NLP models includes a model trained for to determine relevancy to at least one of a particular business or product.

11. The method of claim 2 , wherein the plurality of NLP models includes one or more models trained to at least one of extract entities and identify topics in the deferrable work interaction;

wherein of generating the NLP scores further comprises generating, from the extracted entities or the identified topics, the sparse vector representation of each deferrable work interaction and, based on the sparse vector representation, classifying the deferrable work interaction in accordance with a predefined list of topic categories or complaint categories.

12. The method of claim 2 , wherein the plurality of NLP models includes a model trained to provide a sentiment score classifying the deferrable work interaction as having a sentiment that is positive, negative, neutral, or mixed and a magnitude related thereto.

13. The method of claim 2 , further comprising the steps of:

for each of the deferrable work interactions, receiving from the agent assigned to handle the response, an agent provided priority score; and

using a comparison between the agent provided priority score and the priority score of the deferrable work interaction to update a training of the priority model.

14. The method of claim 2 , further comprising the step of:

sorting the deferrable work interactions according to a relative value of the priority scores;

wherein the factors that the optimization process considers further includes an order of the deferrable work interactions given the sorting.

15. A system related to optimizing a workflow in a contact center in which deferrable work interactions are prioritized and assigned to agents for handling, the system comprising:

a processor; and

a memory, wherein the memory stores instructions that, when executed by the processor, cause the processor to perform the steps of:

providing:

a plurality of natural language processing (NLP) models, each NLP model being configured to accept text from a given deferrable work interaction as an input and generate NLP scores indicating how the given deferrable work interaction rates in accordance with a characteristic, wherein to generate the NLP scores of the given deferrable work interaction, each NLP model generates a sparse vector representation of the given deferrable work interaction;

a priority model configured to accept as inputs the scores generated from the plurality of NLP models and generate a priority score related to a priority characteristic indicating how the given deferrable work interaction should be prioritized for handling relative to other ones of the deferrable work interactions;

receiving the deferrable work interactions;

using text derived from the deferrable working interactions as inputs to the plurality of NLP models to generate the NLP scores for each of the deferrable work interactions;

using the generated NLP scores as inputs to the priority model to generate the priority score for each of the deferrable work interactions;

using the generated NLP scores to identify one or more candidate agents of the agents for handling each of the deferrable work interactions;

receiving an inbound work forecast for the contact center that predicts inbound work levels expected over one or more future work periods;

receiving, in relation to the one or more future work periods, agent work schedule data describing anticipated work schedules for the agents;

using an optimization process to generate an optimized workflow for the deferrable work interactions, wherein, for each of the deferrable work interactions, the optimized workflow comprises an assignment in which a selected agent is selected from the candidate agents for handling the deferrable work interaction and a target timeframe is scheduled for handling of the deferrable work interaction; and

routing each of the deferrable work interactions in accordance with the assignments of the optimized workflow;

wherein the optimization process is configured to optimize according to the following factors:

the priority score generated for each of the deferrable work interactions;

an expected availability over the one or more future work periods of the one or more candidate agents identified for each of the deferrable work interactions as determined from:

the agent work schedule data over the one or more future work periods; and

the predicted inbound work levels over the one or more future work periods given the inbound work forecast.

16. The system of claim 15 , wherein the deferrable work interactions each comprises an email communication sent by a customer that requests a response from the contact center;

wherein the memory further stores instructions that, when executed by the processor, cause the processor to perform the step of:

determining, for each of the deferrable work interactions, a response deadline indicating a threshold for responding to the deferrable work interaction;

determining, for each of the deferrable work interactions, a negative economic impact to the contact center for not responding to the deferrable work interaction before the threshold;

wherein the factors that the optimization process considers further includes the response deadline determined for each of the deferrable work interactions and the negative economic impact determined for each of the deferrable work interactions.

17. The system of claim 15 , wherein the priority model comprises:

an urgency component that provides an urgency score; and

an importance component that provides an importance score;

wherein the priority score is derived from a weighted combination of both the urgency score and the importance score.

18. The system of claim 15 , wherein the inbound work forecast comprises predictions as to an inbound work levels of immediate work interactions expected over the one or more future work periods; and

wherein an immediate work interaction is defined as a synchronous interaction involving a real-time exchange of text or voice between one of the agents and a customer.

19. The system of claim 15 , wherein the step of using the generated NLP scores to identify the one or more candidate agents for handling each of the deferrable work interactions includes:

identifying preferred agent characteristics for each of the deferrable work interactions; and

comparing the identified preferred agent characteristics against actual agent characteristics of the agents and determining the one or more candidate agents as being ones of the agents having a most favorable match;

wherein the preferred agent characteristics are defined as being agent characteristics found to more likely produce a favorable outcome given a characteristic of a given deferrable work interaction.

20. The system of claim 15 , wherein the plurality of NLP models includes a model trained for to determine relevancy to at least one of a particular business or product;

wherein the plurality of NLP models includes one or more models trained to at least one of extract entities and identify topics in the deferrable work interaction; and

wherein the plurality of NLP models includes a model trained to provide a sentiment score classifying the deferrable work interaction as having a sentiment that is positive, negative, neutral, or mixed and a magnitude related thereto.