i have been trying to formulate something that can provide us a clear picture of how and when the call back is made, i have confiigured an EWT at ICM end and passed it to the application but thats not exactly what its using, i believe cvp has its own ewt and DQrate that it relies on, i am not able to figure out what parameters & formula does it considers.
because i am statically setting the EWT to 5 minutes it makes a callback call after 2 minutes or so and sometimes less, even when i try to figure out the calculation via below formula it still does not match, when i look at the Servlet page of cvp, DQRATE is equal to the EWT but remaining time drops very fast.
The process for determining callback time is as follows:
The Average Dequeue Time (D) is calculated using the formula, D = (EWT + F)/N
EWT is the estimated wait time for a new Courtesy Callback call.
F is the number of seconds that the first call is already in position in the queue.
N is the number of calls in queue.
The remaining time for all Courtesy Callback calls in the queue is updated using the formula: R(p) = p*D - F - C:
p = 1, ..., N
R(p) is the remaining time for the pth queue position Courtesy Callback call.C, the post-callback time, is the sum of the time it takes to retrieve the Courtesy Callback caller back on the phone and the SLA time.
Secondly, when an EWT is sent to the application, the agents availability would be considered but does it considers agent availability on the Queue while dialing the call out in any way?
From ICM we send QueueApp, is that the name of the virtual Queue that it will maintain or it will create a virtual Q of that skillgroup on which the call is Queued?
You didn't mention the version you are on, but you likely will find that due to some defects in some versions, that you'll likely basically be using a statically passed EWT. If you do a search of bugs, you'll find some of the additional steps that you need to do, like this one.
Thank you for your response, I am currently using 10.5 version of ucce as well as cvp.
I just wanted to confirm
1. EWT is passed from ICM and afterwards cvp calculates the EWT on regular intervals?
2. Does CVP considers agent availability while calculating its own EWT?
3.what Formula does CVP uses while calculating its EWT, is it the same one i mentioned above?
4.what formula does cvp use to calculate the DQrate?
5. What virtual Q does it uses, is it the Skillgroup on which the call is Queued or it would be the QueueApp that we have passed to the CCB application?
6. Is it possible that we can pass ASA or AHT and PIQ to the CVP, so it can consider them as well, while calculating the remaining time in Queue?
Unfortunately you'll probably have to do a search to see what other CCB defects are applicable to your specific version of CVP, including what ES you have applied.
But bottom line, there's no direct link between agent availability and the EWT. So you can, under certain circumstances, wind up with a call in queue that is waiting to reach 0 for the EWT countdown, while you have an agent available.