AT&T DEFINITY Generic 3 Manuale Utente
Advanced Vector Routing
6-2
Issue 4 September 1995
Expected Wait Time (EWT)
EWT Routing allows you to make routing decisions based on the time that a caller
can expect to wait in queue. This wait time can be predicted for a split or for a
call. When predicted for a split, the wait time indicates the amount of time the
caller can expect to wait if the call is queued to the specified split. When
predicted for a call, the wait time indicates the time remaining that the caller can
expect to wait in queue until the call is serviced from the queue. The expected
wait time can also be passed to a VRU so that a caller can be notified of his or
her expected time in queue. The
can expect to wait in queue. This wait time can be predicted for a split or for a
call. When predicted for a split, the wait time indicates the amount of time the
caller can expect to wait if the call is queued to the specified split. When
predicted for a call, the wait time indicates the time remaining that the caller can
expect to wait in queue until the call is serviced from the queue. The expected
wait time can also be passed to a VRU so that a caller can be notified of his or
her expected time in queue. The
expected-wait
conditional can be used with
either the
goto
or
check-backup
commands.
Call vectoring offers several conditionals that can be used to estimate the time a
caller will be delayed waiting in queue, for example, EWT, rolling ASA and Oldest
Call Waiting (OCW). EWT is the most accurate of these conditionals. It takes into
account more real-time and historical information than the other predictors. For
example, priority level, position in queue, number of working agents, etc.
caller will be delayed waiting in queue, for example, EWT, rolling ASA and Oldest
Call Waiting (OCW). EWT is the most accurate of these conditionals. It takes into
account more real-time and historical information than the other predictors. For
example, priority level, position in queue, number of working agents, etc.
EWT is very responsive to changing call center conditions. For example, it
adjusts instantly to any staffing changes in the split; if an agent moves into or out
of auxiliary work mode, the wait time predictions adjust immediately.
adjusts instantly to any staffing changes in the split; if an agent moves into or out
of auxiliary work mode, the wait time predictions adjust immediately.
EWT does not include the time in a call vector before the call enters a queue. It
also does not include the time the call rings at a voice terminal after it is removed
from the queue.
also does not include the time the call rings at a voice terminal after it is removed
from the queue.
See "When to Use Wait Time Predictions" later in this chapter for a description of
when the predictions are most accurate and the circumstances that will limit their
accuracy.
when the predictions are most accurate and the circumstances that will limit their
accuracy.
EWT for a Split
The EWT for a split is the time that a new call would be expected to remain in
queue if it were queued to the split at the specified priority level. It is generally
used to determine if a call should be queued to the split.
queue if it were queued to the split at the specified priority level. It is generally
used to determine if a call should be queued to the split.
For example, the following vector uses EWT for a split to determine if a call
should be queued to that split.
should be queued to that split.
Figure 6-1.
EWT for a Split
If there are agents available, EWT is zero.
1. goto step 3 if expected-wait for split 1 pri l < 600
2. busy
3. queue-to main split 1 pri l
4. announcement 3001
5. wait-time 998 secs hearing music
2. busy
3. queue-to main split 1 pri l
4. announcement 3001
5. wait-time 998 secs hearing music