Hello everybody,
We had working on this particular problem for a quite a time now.We have been researching on this particular problem to see as which parameters hypotheses the customers irritation level in case of time lost in queqe.Though lot of parameters have been found relevant a definte answers is still to be found.
Once the parameters are found we will use AI and Other Linguistic Techniques to alter responses at different levels of time intervals.This would indeed increase the level of customer comfortability in such situations.
I would like to put forward a simple example for the above mentioned.
suppose there are 5 people in the ACD and all of them are being played the same message.Assumption is we have a FIFO systems in place ,it is most likely that the person who has been in the queqe for the most time may hear a repeat message.Now its here wherein we can use AI as a solution.Intelligent computing helps us to deal with a customer in a more Human interactive way.
But for that we need to know more as what parameters are correlated with this problem.According to time spent in the queqe we can have diffrent voice messages.
We would like to estimate (w,Tw, q,Tq), waiting items, waiting time, queued items, and queuing time, mean and standard deviation of each.
Assumptions made about queuing model is summarized using a notation:
G – general distribution, M – exponential distribution, D – deterministic fixed rate
Accordingly models are named: M/M/1 refers to Poisson arrival , exponential service .
If anybody of you have done any further research on the above mentioned please tell me.
Thanx
sudip |