PREDICTION OF THRESHOLD EXCEPTIONS BASED ON REAL TIME OPERATING INFORMATION

• US 20100274637A1
• Filed: 04/23/2009
• Published: 10/28/2010
• Est. Priority Date: 04/23/2009
• Status: Active Grant

• Current Assignee: Avaya Incorporated
• Original Assignee: Avaya Incorporated
• Inventors: Smith, Lynn R., Li, Hongguang, Sobus, Katherine A., Paddock, Henry R.
• Granted Patent: US 8,175,254 B2
• Time in Patent Office: Days
• US Class Current: 705/11
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/5009 Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
• CPC Class Codes: H04L 41/5025 by proactively reacting to service quality change, e.g. by reconfiguration after service quality degradation or upgrade
• CPC Class Codes: H04L 43/16 Threshold monitoring

Abstract

A contact center includes a real time threshold prediction module that determines that a selected performance measure will, during a future time interval, likely cross a selected threshold and, in response, performs at least one of the following operations: (i) determines, for the likely crossing of the selected threshold, a consequence in a selected service level agreement and determines a set of corrective actions to reduce a likelihood that the selected performance measure will cross the selected threshold.

Claims

1. In a contact center comprising a plurality of agents servicing work items from a plurality of work item queues, a method comprising:

(a) determining, by a processor, that a selected performance measure will, during a future time interval, likely cross a selected threshold;

(b) in response, performing, by a processor, at least one of the following steps;

(B1) determining a financial impact of crossing the selected threshold; and

(B2) determining a set of correction actions to reduce a likelihood that the selected performance measure will cross the selected threshold.

2. The method of claim 1, wherein the performance measure is one or more of service level, expected wait time, predicted wait time, actual wait time, number of contacts waiting, percent abandons, average active during a call, average speed of answer, average hold time, percent of work items that are one and done, match rate, Percent Agent Disconnects First, Percent available, Percent Conferences, Percent Discussion from On Hold, Percent in Service Level, Percent Occupancy, Percent Redirects, Percent Transfers, Percent Transfers to Same Routing Point, Abandon Duration, Abandons, Abandons, Percent, Acceptables, Accepts, Active Duration, Active Duration, Average, Active Staffed Agents by Routing Point, Active Staffed Agents by Queue, Actives, Adherence to schedule, Adherence to procedures, Adjunct Attempts, Adjunct Redirects, Agent Occupancy, Agent State Duration, Agents Staffed, Agents in Aux. State, Agent Turnover rate, Alert Duration, Alert Duration, Average, Alerting Agents Staffed, Arrivals, Audio Difficulties, Auto Reserve Duration, Average Speed of Answer (ASA), Average Handle Time (AHT), Behind Target Duration, Behind Target Not Reserving Duration, Busies, Busies, Percent, Call Blockage, Cancels, Cancels, Percent, Conferences, Contacts at Agent, Contacts Finished Waiting, Contacts Waiting, Cost per contact (per media), Customer defection rate, Customer Satisfaction, Critical Duration, Curr. Max. Wait Duration, Dequeues, Direct Agent Contacts Waiting, Discussion from On Hold, Error or rework percentage or volume, Exp. Wait, First Call Resolution (First resolution rate), Handle Duration, Handle Duration, Average, Handles, Held Contacts, Idle Agents Staffed, Idle Duration, In Focus Hold Agents Staffed, In Focus Hold Duration, Inflows, Initiating Agents Staffed, Interflows, Initiate Duration, Interactions, Interactions at Agent, Max Delay in Queue (longest delay in queue), Net. Discussion, Net. in Duration, Net. Polls, Net. Redirects, Non-Working Aux. Agents Staffed, Non-Working Aux. Duration, Number of times a customer contacted the company, Occupied Duration, On Hold Agents Staffed, On Hold Contacts, On Hold Duration, On Target Duration, Out of Focus Hold Duration, Outflows, Outflows, Percent, Preview Agents Staffed, Preview Duration, Previews, Quality Monitoring Scores, Queue Duration, Redirects, Reserve Handles, Reserve Standby Agents Count, Returns, Service Level, Self Service Availability, Short Contacts, Short Seizures, Silence Duration, Staff Shrinkage (Shrinkage), State Duration, Staffed Duration, Successful Sales (number) or percentage, Successful upsell (number or percentage), Transfer Rate, Unknown Agents Staffed, Unknown Duration, Unoccupied Duration, Unstaffed Queue, Wait Duration, Working Aux. Agents Staffed, Working Aux. Duration, Wrap-up Agents Staffed, Wrap-up Duration, Wrap-up Duration, Average, and Wrap-ups, and any other performance indicators known or later developed.

3. The method of claim 1, wherein step (a) comprises:

(A1) determining, from substantially real-time contact center operating information and by a processor, a current value for the selected performance measure;

(A2) determining, from historic contact center operating information and by a processor, a set of historic values for the selected performance measure;

(A3) determining, by regression analysis and by a processor, at least one of a mathematical function and a curve that approximately characterizes the historic and current values for the selected performance measure; and

(A4) determining, by a processor and based on the at least one of a mathematical function and curve, that the selected performance measure will cross the selected threshold and an estimate of when the selected performance measure will cross the threshold.

4. The method of claim 3, further comprising:

(c) determining, by a processor, one or more levels of confidence that at least one of a mathematical function and curve correctly characterizes the historic and current selected performance measure values, that the selected performance threshold will cross the threshold, and/or that the estimate of when the selected performance measure is/are correct.

5. The method of claim 3, wherein (i) is performed and fuirther comprising:

(d) determining that the one or more levels of confidence indicates that a threshold crossing is likely; and

(e) in response to step (d), assigning a value to the threshold crossing event, wherein the value is a function of a penalty under a service level agreement between a contact center operator and a client.

6. The method of claim 4, wherein step (ii) is performed and further comprising:

(f) determining that the event value is greater than a selected value; and

(g) in response to step (f), determining a set of recommendations to reduce a likelihood that the threshold crossing event will occur.

7. The method of claim 5, wherein each recommendation in the set of recommendations has a corresponding at least one of a cost and benefit and a likelihood of success.

8. A computer readable medium comprising processor executable instructions to perform the steps of claim 1.

9. A contact center, comprising:

(a) a plurality of work item queues;

(b) a plurality of agents servicing work items from the plurality of work item queues; and

;

(c) a real time threshold prediction module operable to;

(C1) determine that a selected performance measure will, during a future time interval, likely cross a selected threshold;

(C2) in response, perform at least one of the following operations;

(i) determine a consequence of crossing the selected threshold; and

(ii) determine a set of corrective actions to reduce a likelihood that the selected performance measure will cross the selected threshold.

10. The contact center of claim 9, wherein the performance measure is one or more of service level, expected wait time, predicted wait time, actual wait time, number of contacts waiting, percent abandons, average active during a call, average speed of answer, average hold time, percent of work items that are one and done, match rate, Percent Agent Disconnects First, Percent available, Percent Conferences, Percent Discussion from On Hold, Percent in Service Level, Percent Occupancy, Percent Redirects, Percent Transfers, Percent Transfers to Same Routing Point, Abandon Duration, Abandons, Abandons, Percent, Acceptables, Accepts, Active Duration, Active Duration, Average, Active Staffed Agents by Routing Point, Active Staffed Agents by Queue, Actives, Adherence to schedule, Adherence to procedures, Adjunct Attempts, Adjunct Redirects, Agent Occupancy, Agent State Duration, Agents Staffed, Agents in Aux. State, Agent Turnover rate, Alert Duration, Alert Duration, Average, Alerting Agents Staffed, Arrivals, Audio Difficulties, Auto Reserve Duration, Average Speed of Answer (ASA), Average Handle Time (AHT), Behind Target Duration, Behind Target Not Reserving Duration, Busies, Busies, Percent, Call Blockage, Cancels, Cancels, Percent, Conferences, Contacts at Agent, Contacts Finished Waiting, Contacts Waiting, Cost per contact (per media), Customer defection rate, Customer Satisfaction, Critical Duration, Curr. Max. Wait Duration, Dequeues, Direct Agent Contacts Waiting, Discussion from On Hold, Error or rework percentage or volume, Exp. Wait, First Call Resolution (First resolution rate), Handle Duration, Handle Duration, Average, Handles, Held Contacts, Idle Agents Staffed, Idle Duration, In Focus Hold Agents Staffed, In Focus Hold Duration, Inflows, Initiating Agents Staffed, Interflows, Initiate Duration, Interactions, Interactions at Agent, Max Delay in Queue (longest delay in queue), Net. Discussion, Net. in Duration, Net. Polls, Net. Redirects, Non-Working Aux. Agents Staffed, Non-Working Aux. Duration, Number of times a customer contacted the company, Occupied Duration, On Hold Agents Staffed, On Hold Contacts, On Hold Duration, On Target Duration, Out of Focus Hold Duration, Outflows, Outflows, Percent, Preview Agents Staffed, Preview Duration, Previews, Quality Monitoring Scores, Queue Duration, Redirects, Reserve Handles, Reserve Standby Agents Count, Returns, Service Level, Self Service Availability, Short Contacts, Short Seizures, Silence Duration, Staff Shrinkage (Shrinkage), State Duration, Staffed Duration, Successful Sales (number) or percentage, Successful upsell (number or percentage), Transfer Rate, Unknown Agents Staffed, Unknown Duration, Unoccupied Duration, Unstaffed Queue, Wait Duration, Working Aux. Agents Staffed, Working Aux. Duration, Wrap-up Agents Staffed, Wrap-up Duration, Wrap-up Duration, Average, and Wrap-ups.

11. The contact center of claim 9, wherein operation (C1) comprises the sub-operations:

determine, from substantially real-time contact center operating information and by a processor, a current value for the selected performance measure;

determine, from historic contact center operating information, a set of historic values for the selected performance measure;

determine, by regression analysis, at least one of a mathematical function and a curve that approximately characterizes the historic and current values for the selected performance measure; and

determine, based on the at least one of a mathematical function and curve, that the selected performance measure will cross the selected threshold and an estimate of when the selected performance measure will cross the threshold.

12. The contact center of claim 11, wherein the module (C3) determines one or more levels of confidence that the at least one of a mathematical function and curve correctly characterizes the historic and current selected performance measure values, that the selected performance threshold will cross the threshold, and/or that the estimate of when the selected performance measure is/are correct.

13. The contact center of claim 12, wherein the consequence is a financial impact and wherein the module performs (i) the further operations:

(C4) determining that the one or more levels of confidence indicates that a threshold crossing is likely; and

(C5) in response to operation (C4), assigning a value to the threshold crossing event, wherein the value is a function of a penalty under a service level agreement between a contact center operator and a client.

14. The contact center of claim 13, wherein the module performs operation (ii) and the further operations:

(C6) determining that the event value is greater than a selected value; and

(C7) in response to step (C6), determining a set of recommendations to reduce a likelihood that the threshold crossing event will occur.

15. The contact center of claim 14, wherein each recommendation in the set of recommendations has a corresponding at least one of a cost and benefit and a likelihood of success.

16. A contact center, comprising:

(a) a plurality of work item queues;

(b) a plurality of agent queues for servicing work items from the plurality of work item queues, wherein a first work item queue corresponds to a first client having a first service level agreement or any other performance measurement agreement with an operator of the contact center and a second work item queue corresponds to a second client having a second service level agreement or any other performance measurement agreement with the contact center operator, wherein the first and second service level or any other performance measurement agreement agreements set forth one or more performance thresholds and, for each such threshold, a consequence in the event that a performance indicator crosses the threshold; and

;

I a real time threshold prediction module operable to;

(C1) determine that a selected performance measure will, during a future time interval, likely cross a selected threshold;

(C2) in response, perform at least one of the following operations;

(i) determine, for the likely crossing of the selected threshold, a financial penalty in a selected service level agreement or any other performance measurement agreement; and

(ii) determine a set of corrective actions to reduce a likelihood that the selected performance measure will cross the selected threshold.

17. The contact center of claim 16, wherein the performance measure is one or more of service level or any other performance measurement agreement, expected wait time, predicted wait time, actual wait time, number of contacts waiting, percent abandons, average active during a call, average speed of answer, average hold time, percent of work items that are one and done, match rate, Percent Agent Disconnects First, Percent available, Percent Conferences, Percent Discussion from On Hold, Percent in Service Level, Percent Occupancy, Percent Redirects, Percent Transfers, Percent Transfers to Same Routing Point, Abandon Duration, Abandons, Abandons, Percent, Acceptables, Accepts, Active Duration, Active Duration, Average, Active Staffed Agents by Routing Point, Active Staffed Agents by Queue, Actives, Adherence to schedule, Adherence to procedures, Adjunct Attempts, Adjunct Redirects, Agent Occupancy, Agent State Duration, Agents Staffed, Agents in Aux. State, Agent Turnover rate, Alert Duration, Alert Duration, Average, Alerting Agents Staffed, Arrivals, Audio Difficulties, Auto Reserve Duration, Average Speed of Answer (ASA), Average Handle Time (AHT), Behind Target Duration, Behind Target Not Reserving Duration, Busies, Busies, Percent, Call Blockage, Cancels, Cancels, Percent, Conferences, Contacts at Agent, Contacts Finished Waiting, Contacts Waiting, Cost per contact (per media), Customer defection rate, Customer Satisfaction, Critical Duration, Curr. Max. Wait Duration, Dequeues, Direct Agent Contacts Waiting, Discussion from On Hold, Error or rework percentage or volume, Exp. Wait, First Call Resolution (First resolution rate), Handle Duration, Handle Duration, Average, Handles, Held Contacts, Idle Agents Staffed, Idle Duration, In Focus Hold Agents Staffed, In Focus Hold Duration, Inflows, Initiating Agents Staffed, Interflows, Initiate Duration, Interactions, Interactions at Agent, Max Delay in Queue (longest delay in queue), Net. Discussion, Net. in Duration, Net. Polls, Net. Redirects, Non-Working Aux. Agents Staffed, Non-Working Aux. Duration, Number of times a customer contacted the company, Occupied Duration, On Hold Agents Staffed, On Hold Contacts, On Hold Duration, On Target Duration, Out of Focus Hold Duration, Outflows, Outflows, Percent, Preview Agents Staffed, Preview Duration, Previews, Quality Monitoring Scores, Queue Duration, Redirects, Reserve Handles, Reserve Standby Agents Count, Returns, Service Level, Self Service Availability, Short Contacts, Short Seizures, Silence Duration, Staff Shrinkage (Shrinkage), State Duration, Staffed Duration, Successful Sales (number) or percentage, Successful upsell (number or percentage), Transfer Rate, Unknown Agents Staffed, Unknown Duration, Unoccupied Duration, Unstaffed Queue, Wait Duration, Working Aux. Agents Staffed, Working Aux. Duration, Wrap-up Agents Staffed, Wrap-up Duration, Wrap-up Duration, Average, and Wrap-ups.

18. The contact center of claim 16, wherein operation (C1) comprises the sub-operations:

determine, from substantially real-time contact center operating information and by a processor, a current value for the selected performance measure;

determine, from historic real-time contact center operating information, a set of historic values for the selected performance measure;

determine, by regression analysis, at least one of a mathematical function and a curve that approximately characterizes the historic and current values for the selected performance measure; and

determine, based on the at least one of a mathematical function and curve, that the selected performance measure will cross the selected threshold and an estimate of when the selected performance measure will cross the threshold and wherein the module (C3) determines one or more levels of confidence that the at least one of a mathematical function and curve correctly characterizes the historic and current selected performance measure values, that the selected performance threshold will cross the threshold, and/or that the estimate of when the selected performance measure is/are correct.

19. The contact center of claim 18, wherein the module performs (i) and the further operations:

(C4) determining that the one or more levels of confidence indicates that a threshold crossing is likely; and

(C5) in response to operation (C4), assigning a value to the threshold crossing event, wherein the value is a function of a penalty under a service level agreement between a contact center operator and a client.

20. The contact center of claim 19, wherein step (ii) is performed and further comprising:

(f) determining that the event value is greater than a selected value; and

(g) in response to step (f), determining a set of recommendations to reduce a likelihood that the threshold crossing event will occur and wherein each recommendation in the set of recommendations has a corresponding at least one of a cost and benefit and a likelihood of success.