Techniques for estimating expected performance in a task assignment system
View Patent ↗Techniques for estimating expected performance of a task assignment strategy in a task assignment system are disclosed. In one particular embodiment, the techniques may be realized as a method comprising receiving, by at least one computer processor communicatively coupled to a task assignment system, a plurality of historical agent task assignments; determining, by the at least one computer processor, a sample of the plurality based on a strategy for pairing agents with tasks; determining, by the at least one computer processor, an expected performance of the strategy based on the sample; outputting, by the at least one computer processor, the expected performance; and optimizing, by the at least one computer processor, the performance of the task assignment system based on the expected performance.
1. A method for validating a preferred task assignment strategy in a task assignment system comprising:
receiving, by at least one computer processor communicatively coupled to and configured to perform task assignment operations in the task assignment system, a first plurality of historical agent-task assignments;
determining, by the at least one computer processor, a baseline performance level of the task assignment system using an underlying task assignment strategy based on each outcome of the first plurality of historical agent-task assignments;
determining, by the at least one computer processor, a first subset of historical agent-task assignments by excluding any of the first plurality of historical agent-task assignments that exceed a threshold closeness of fit to the preferred task assignment strategy;
determining, by the at least one computer processor, an estimated performance level of the task assignment system using the preferred task assignment strategy based on each outcome of the first subset of historical agent-task assignments;
outputting, by the at least one computer processor, an expected gain attributable to using the preferred task assignment strategy in the task assignment system by comparing the estimated performance level to the baseline performance level; and
establishing, in a switch of the task assignment system, a connection between a task and an agent based upon the preferred task assignment strategy.
2. The method of claim 1 , wherein the task assignment system is a contact center system, and wherein each of the first plurality of historical agent-task assignments is a historical agent-contact assignment.
3. The method of claim 1 , wherein the preferred task assignment strategy is a behavioral pairing strategy.
4. The method of claim 1 , further comprising:
determining, by the at least one computer processor, that the first plurality of historical agent-task assignments are approximately uniformly distributed throughout the possible pairing space, wherein the underlying task assignment strategy is a FIFO task assignment strategy.
5. The method of claim 1 , further comprising:
determining, by the at least one computer processor, that the first plurality of historical agent-task assignments are clustered around agents having higher performance rankings, wherein the underlying task assignment strategy is a performance-based task assignment strategy.
6. The method of claim 5 , further comprising:
reweighting the first subset of historical agent-task assignments to remove bias introduced by the performance-based task assignment strategy, wherein the estimated performance level and the expected gain are based on the reweighted first subset of historical agent-task assignments.
7. The method of claim 1 , further comprising:
determining, by the at least one computer processor, the threshold closeness of fit based on a corresponding historical average amount of choice for agent-task pairing in the task assignment system.
8. A system for validating a preferred task assignment strategy in a task assignment system comprising:
at least one computer processor communicatively coupled to and configured to perform task assignment operations in the task assignment system, wherein the at least one computer processor is further configured to:
receive by at least one computer processor communicatively coupled to and configured to perform task assignment operations in the task assignment system, a first plurality of historical agent-task assignments;
determine a baseline performance level of the task assignment system using an underlying task assignment strategy based on each outcome of the first plurality of historical agent-task assignments;
determine a first subset of historical agent-task assignments by excluding any of the first plurality of historical agent-task assignments that exceed a threshold closeness of fit to the preferred task assignment strategy;
determine an estimated performance level of the task assignment system using the preferred task assignment strategy based on each outcome of the first subset of historical agent-task assignments;
output an expected gain attributable to using the preferred task assignment strategy in the task assignment system by comparing the estimated performance level to the baseline performance level; and
establish, in a switch of the task assignment system, a connection between a task and an agent based upon the preferred task assignment strategy.
9. The system of claim 8 , wherein the task assignment system is a contact center system, and wherein each of the first plurality of historical agent-task assignments is a historical agent-contact assignment.
10. The system of claim 8 , wherein the preferred task assignment strategy is a behavioral pairing strategy.
11. The system of claim 8 , wherein the at least one computer processor is further configured to:
determine that the first plurality of historical agent-task assignments are approximately uniformly distributed throughout the possible pairing space, wherein the underlying task assignment strategy is a FIFO task assignment strategy.
12. The system of claim 8 , wherein the at least one computer processor is further configured to:
determine that the first plurality of historical agent-task assignments are clustered around agents having higher performance rankings, wherein the underlying task assignment strategy is a performance-based task assignment strategy.
13. The system of claim 12 , wherein the at least one computer processor is further configured to:
reweight the first subset of historical agent-task assignments to remove bias introduced by the performance-based task assignment strategy, wherein the estimated performance level and the expected gain are based on the reweighted first subset of historical agent-task assignments.
14. The system of claim 8 , wherein the at least one computer processor is further configured to:
determine the threshold closeness of fit based on a corresponding historical average amount of choice for agent-task pairing in the task assignment system.
15. An article of manufacture for validating a preferred task assignment strategy in a task assignment system comprising:
a non-transitory processor readable medium; and
instructions stored on the medium;
wherein the instructions are configured to be readable from the medium by at least one computer processor communicatively coupled to and configured to perform task assignment operations in the task assignment system and thereby cause the at least one computer processor to operate so as to:
receive by at least one computer processor communicatively coupled to and configured to perform task assignment operations in the task assignment system, a first plurality of historical agent-task assignments;
determine a baseline performance level of the task assignment system using an underlying task assignment strategy based on each outcome of the first plurality of historical agent-task assignments;
determine a first subset of historical agent-task assignments by excluding any of the first plurality of historical agent-task assignments that exceed a threshold closeness of fit to the preferred task assignment strategy;
determine an estimated performance level of the task assignment system using the preferred task assignment strategy based on each outcome of the first subset of historical agent-task assignments;
output an expected gain attributable to using the preferred task assignment strategy in the task assignment system by comparing the estimated performance level to the baseline performance level; and
establish, in a switch of the task assignment system, a connection between a task and an agent based upon the preferred task assignment strategy.
16. The article of manufacture of claim 15 , wherein the task assignment system is a contact center system, and wherein each of the first plurality of historical agent-task assignments is a historical agent-contact assignment.
17. The article of manufacture of claim 15 , wherein the preferred task assignment strategy is a behavioral pairing strategy.
18. The article of manufacture of claim 15 , wherein the at least one computer processor is further caused to operate so as to:
determine that the first plurality of historical agent-task assignments are approximately uniformly distributed throughout the possible pairing space, wherein the underlying task assignment strategy is a FIFO task assignment strategy.
19. The article of manufacture of claim 15 , wherein the at least one computer processor is further caused to operate so as to:
determine that the first plurality of historical agent-task assignments are clustered around agents having higher performance rankings, wherein the underlying task assignment strategy is a performance-based task assignment strategy.
20. The article of manufacture of claim 19 , wherein the at least one computer processor is further caused to operate so as to:
reweight the first subset of historical agent-task assignments to remove bias introduced by the performance-based task assignment strategy, wherein the estimated performance level and the expected gain are based on the reweighted first subset of historical agent-task assignments.
21. The article of manufacture of claim 15 , wherein the at least one computer processor is further caused to operate so as to:
determine the threshold closeness of fit based on a corresponding historical average amount of choice for agent-task pairing in the task assignment system.