Telephonic communications – Centralized switching system – Call distribution to operator
Reexamination Certificate
2000-08-10
2003-10-28
Matar, Ahmad F. (Department: 2642)
Telephonic communications
Centralized switching system
Call distribution to operator
C379S265140, C379S265110, C379S265010, C379S265020
Reexamination Certificate
active
06639982
ABSTRACT:
TECHNICAL FIELD
This invention relates generally to telecommunications systems and more specifically, to methods and systems for dynamically forcing and distributing calls to agents within large team call servicing centers.
BACKGROUND ART
Force management systems for large team call servicing centers are well known in the prior art. These systems forecast the required number of call handling agents to handle the anticipated call volume as well as distribute calls to the call handling agents in some predetermined manner.
In determining the required number of agents to handle the anticipated call volume, call centers rely primarily on two variables: the estimated, or offered, load and grade of service. The offered load is a unit of measurement reflecting the estimated total work time of the incoming call traffic and is calculated by multiplying the estimated number of calls within a certain time period by the estimated average work time for those calls. In estimating the number of calls, current force management systems typically utilize time-series analyses of historic data as well as current telephone switch monitored data to forecast the number of incoming calls for a particular time period. Once the offered load is calculated, the number of agents required to service this specific load is determined. Typically, further adjustments are made to the number of agents working within a particular period of time as force managers evaluate the previous time period's results as well as the trends for the day. In U.S. Pat. No. 5,185,780 to Leggett a method for predicting the required number of agents required to provide a given service level in a force management system is taught. Under this method, the force management system generates call handling performance data from which an offered load is calculated. Using this offered load value, the method calculates predictor values and uses these values in successive Erlang C calculations to locate the desired number of agents required to provide a given service level.
As the Leggett invention teaches, present call center force management systems attempt to match the appropriate number of agents with the estimated offered load to ensure that some predetermined acceptable level of service is achieved. There are many parameters by which the level, or grade, of service to the calling customer may be evaluated. However, the two most typical measurements include average delay in queue (“average delay”) and percent abandon. Average delay refers to the length of time a caller must wait in queue before the caller is connected to an agent position. Percent abandon refers to the percentage of callers who abandon their efforts to seek agent assistance. Understandably, there is a very strong correlation between average delay and percent abandon.
Another aspect of the agent staffing function that critically affects the ability to manage service levels is the agent work force configuration. Current state of the art utilizes a single, large team approach that allows any incoming call to reach any call agent Under this approach, the offered load must be calculated for each distinct period of time and management must schedule a corresponding single, large team of agents to meet this offered load while providing a predetermined service level. An example of the range of agent team size necessary to try to meet the offered load demonstrates the wide variability in team size throughout a twenty-four hour period. In one specific type of large team call center, telephone directory assistance, team sizes may range from as few as 30 call agents in the early morning hours to as many as 250 agents and more during the peak busy hours.
Under the single, large team methodology, management may need to make adjustments to the large agent team when service reaches some critical level by scheduling one or more additional agents. For example, when the offered load exceeds the carried load for a particular increment of time, e.g. 15 minutes, 30 minutes, etc., the result is a degradation in service. When service is degraded, the average delay and the percent abandons both increase in some concomitant manner. This percent abandon value may be as high as 20-50% for some quarter hour intervals. Typically, management this agent workforce adjustment by monitoring and evaluating the results of a particular time period for a service degradation, deciding how best to manage the service degradation, and finally arranging for additional agent resources, if needed, by calling in additional agents or offering overtime to those agents presently on duty. It is important to note that no existing force management system is capable of making dynamic adjustments to agent team size in real-time.
In addition to estimating the required number of agents, current force management systems distribute calls to the agents in some predetermined manner. Typically, calls are distributed by an automatic call distributor (ACD) to one of a plurality of call agents based on some predetermined notion of “fairness.” For example, in U.S. Pat. No. 5,828,747 to Fisher, et al. a call distribution methodology is taught based on an individual agent's occupancy. Call occupancy refers to the amount of time that an agent spends on actually handling calls and is typically expressed as a percentage of the total amount of time the agent has been available to handle a call. Under the Fisher methodology, individual agent occupancies are calculated and a queue of available agents is ordered in the inverse order of the agents occupancies. Calls are distributed to the agents in the inverse order of their individual agent occupancies specifically to improve the equity of the call distribution.
Other call distribution methodologies based on some notion of “fairness” include the distribution of calls to an agent who has been idle the longest or the distribution of calls to an agent to ensure that all agents handle an equal number of calls. Idle time re-refers to that period of time in which the agent is not handling a call or performing a call related task. Lastly, other call distribution methodologies include distributing calls based on a caller's originating number or geographic location.
Current force management systems may also attempt to perform the agent forcing and call distribution function in some cost-optimizing manner. Call centers often attempt to strike a balance between satisfying the grade of service objectives and minimizing the agent variable costs. When the offered load exceeds the carried load, the call volume that is actually being handled by the call agents, service is degraded and customers abandon their efforts to seek assistance from an agent. The percent abandon increases and service may be significantly degraded. Moreover, this increase in the number of callers who abandon their calling efforts results in a portion of these same callers attempting to seek agent assistance again. These regenerated attempts are not random and cause the offered load to further increase exponentially, thereby exacerbating the percent abandon service problem. Conversely, when the offered load is less than the carried load, agents are idle, resulting in an inefficient use of labor. Neither of these conditions results in optimal performance. An increasing percentage of abandoned calls indicates customer dissatisfaction with the company; correspondingly, reducing the percentage of abandoned calls by increasing the number of available agents may drive up cost beyond an optimal level.
In addition to minimizing agent cost, current methodologies may attempt to optimize agent utility either by distributing calls to an agent with a particular skill or by utilizing an agent's idle time to perform tasks other than the handling of calls. U.S. Pat. No. 5,206,903 to Kohler et al. teaches an automatic call distribution in which each incoming call, with each call having been assigned up to three prioritized skill numbers, attempts to be matched with an agent possessing that particular skill. Each agent possesses up to tou
Stuart Robert Oden
Stuart Scott Patterson
Stuart, II Robert Oden
Agdeppa Hector
Liell & McNeil
Matar Ahmad F.
Six Sigma, Inc.
LandOfFree
Method and apparatus for agent forcing and call distribution... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for agent forcing and call distribution..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for agent forcing and call distribution... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3129097