Granted Patent B2

US 12,688,473 · App. 18/615,815 · Granted Jul 21, 2026

Inventor: Denys Liubyvyi (Washington, DC)

Assignee: AFINITI AI LIMITED

G06Q10/063114G06Q10/06312G06Q10/06398

Abstract

Claims (84)

1 . A method for a task assignment system, the method comprising:

determining, by at least one computer processor configured to perform workforce management operations in the task assignment system, that the task assignment system is in a first state, wherein at the time that the task assignment system is determined to be in the first state, there are N 1 number of agents available for pairing and further wherein N 1 is a non-negative integer;

after determining that the task assignment system is in the first state, calculating, by said at least one computer processor,

first value which indicates an estimated result of performing pairings with N Increased number of agents, wherein N Increased is obtained as a result of increasing N 1 ; and

a second value which indicates an estimated result of performing pairings with N Decreased number of agents, wherein N Decreased is obtained as a result of decreasing N 1 ; and

based on the comparison of the first and second values, determining whether to increase or decrease the N 1 number of agents;

storing, in a task queue, information regarding a task waiting to be paired with an available agent; and

establishing a communication link between an agent available for pairing and the task waiting to be paired with an available agent, wherein

establishing the communication link between the agent and the task comprises using a switch of the task assignment system to establish the communication link,

either increasing or decreasing the N 1 number of agents causes the task assignment system to transition from the first state to a second state which is different from the first state,

each of the first and second values is calculated based on:

a profit associated with performing pairings using a certain pairing strategy with the N Increased number of agents or the N Decreased number of agents;

a profit associated with performing pairings without using a certain pairing strategy with the N Increased number of agents or the N Decreased number of agents; and

a cost associated with performing pairings using a certain pairing strategy with the N Increased number of agents or the N Decreased number of agents,

the profit associated with performing pairings using the certain pairing strategy with the N Increased number of agents or the N Decreased number of agents is determined based at least on an expected loss of each of one or more tasks,

the expected loss of each of said one or more tasks is determined using a hazard function, and

the hazard function indicates a likelihood that a task will be lost with respect to a waiting time duration of the task.

2 . The method of claim 1 , wherein the method comprises:

after determining that the task assignment system is in the first state, calculating, by said at least one computer processor, a plurality of values each of which indicates an estimated result of performing pairings with each of candidate numbers of agents, wherein the candidate numbers of agents include N Increased and N Decreased ;

comparing the plurality of values with each other; and

based on the comparison of the plurality of values, selecting one of the candidate numbers of agents.

3 . The method of claim 1 , wherein the first value which indicates the estimated result of performing pairings with the N increased number of agents is calculated based on a profit associated with performing pairings with the N increased number of agents, and the second value which indicates the estimated result of performing pairings with the N Decreased number of agents is calculated based on a profit associated with performing pairings with the N Decreased number of agents.

4 . The method of claim 1 , wherein the certain pairing strategy is a behavioral pairing strategy.

5 . The method of claim 1 , wherein

whether to increase or decrease the N 1 number of agents is determined based on a metric, and

the metric is related to one or more of:

a cost related to increasing the N 1 number of agents;

a benefit related to decreasing the N 1 number of agents;

a cost related to increasing a time duration during which agents are waiting to be paired;

a benefit related to decreasing a time duration during which agents are waiting to be paired;

a cost related to increasing a time duration during which tasks are waiting to be paired; and/or

a benefit related to decreasing a time duration during which tasks are waiting to be paired.

6 . A task assignment system comprising:

at least one memory; and

at least one computer processor configured to perform workforce management operations in the task assignment system, wherein the task assignment system is configured to:

determine that the task assignment system is in a first state, wherein at the time that the task assignment system is determined to be in the first state, there are N 1 number of agents available for pairing and further wherein N 1 is a non-negative integer; and

after determining that the task assignment system is in the first state, calculate:

a first value which indicates an estimated result of performing pairings with N Increased number of agents, wherein N Increased is obtained as a result of increasing N 1 ; and

a second value which indicates an estimated result of performing pairings with N Decreased number of agents, wherein N Decreased is obtained as a result of decreasing N 1 ; and

based on the comparison of the first and second values, determine whether to increase or decrease the N 1 number of agents;

storing, in a task queue, information regarding a task waiting to be paired with an available agent; and

establishing a communication link between an agent available for pairing and the task waiting to be paired with an available agent, wherein

establishing the communication link between the agent and the task comprises using a switch of the task assignment system to establish the communication link, and

either increasing or decreasing the N 1 number of agents causes the task assignment system to transition from the first state to a second state which is different from the first state each of the first and second values is calculated based on:

a profit associated with performing pairings using a certain pairing strategy with the N Increased number of agents or the N Decreased number of agents;

a profit associated with performing pairings without using a certain pairing strategy with the N Increased number of agents or the N Decreased number of agents; and

a cost associated with performing pairings using a certain pairing strategy with the N Increased number of agents or the N Decreased number of agents,

the profit associated with performing pairings using the certain pairing strategy with the N Increased number of agents or the N Decreased number of agents is determined based at least on an expected loss of each of one or more tasks,

the expected loss of each of said one or more tasks is determined using a hazard function, and

the hazard function indicates a likelihood that a task will be lost with respect to a waiting time duration of the task.

7 . The task assignment system of claim 6 , wherein the task assignment system is configured to:

after determining that the task assignment system is in the first state, calculate, by said at least one computer processor, a plurality of values each of which indicates an estimated result of performing pairings with each of candidate numbers of agents, wherein the candidate numbers of agents include N Increased and N Decreased ;

compare the plurality of values with each other; and

based on the comparison of the plurality of values, select one of the candidate numbers of agents.

8 . The task assignment system of claim 6 , wherein the first value which indicates the estimated result of performing pairings with the N increased number of agents is calculated based on a profit associated with performing pairings with the N increased number of agents, and the second value which indicates the estimated result of performing pairings with the N Decreased number of agents is calculated based on a profit associated with performing pairings with the N Decreased number of agents.

9 . The task assignment system of claim 6 , wherein the certain pairing strategy is a behavioral pairing strategy.

10 . The task assignment system of claim 6 , wherein

whether to increase or decrease the N 1 number of agents is determined based on a metric, and

the metric is related to one or more of:

a cost related to increasing the N 1 number of agents;

a benefit related to decreasing the N 1 number of agents;

a cost related to increasing a time duration during which agents are waiting to be paired;

a benefit related to decreasing a time duration during which agents are waiting to be paired;

a cost related to increasing a time duration during which tasks are waiting to be paired; and/or

a benefit related to decreasing a time duration during which tasks are waiting to be paired.

11 . An article of manufacture comprising:

a non-transitory computer 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 configured to perform workforce management operations in a task assignment system, thereby causing the at least one computer processor to operate so as to:

determine that the task assignment system is in a first state, wherein at the time that the task assignment system is determined to be in the first state, there are N 1 number of agents available for pairing and further wherein N 1 is a non-negative integer; and

after determining that the task assignment system is in the first state, calculate:

a first value which indicates an estimated result of performing pairings with N Increased number of agents, wherein N Increased is obtained as a result of increasing N 1 ; and

a second value which indicates an estimated result of performing pairings with N Decreased number of agents, wherein N Decreased is obtained as a result of decreasing N 1 ; and

based on the comparison of the first and second values, determine whether to increase or decrease the N 1 number of agents;

storing, in a task queue, information regarding a task waiting to be paired with an available agent; and

establishing a communication link between an agent available for pairing and the task waiting to be paired with an available agent, wherein

establishing the communication link between the agent and the task comprises using a switch of the task assignment system to establish the communication link, and

either increasing or decreasing the N 1 number of agents causes the task assignment system to transition from the first state to a second state which is different from the first state,

each of the first and second values is calculated based on:

a profit associated with performing pairings using a certain pairing strategy with the N Increased number of agents or the N Decreased number of agents;

a profit associated with performing pairings without using a certain pairing strategy with the N Increased number of agents or the N Decreased number of agents; and

a cost associated with performing pairings using a certain pairing strategy with the N Increased number of agents or the N Decreased number of agents,

the profit associated with performing pairings using the certain pairing strategy with the N Increased number of agents or the N Decreased number of agents is determined based at least on an expected loss of each of one or more tasks,

the expected loss of each of said one or more tasks is determined using a hazard function, and

the hazard function indicates a likelihood that a task will be lost with respect to a waiting time duration of the task.

Assignments (5)

Continuity (3)

References Cited (355)