Trunk Performance Report; Algorithm For Trunk Performance - Lucent Technologies CentreVu Release 3 Version 5 Forecast User Manual

How the Forecast System Generates Data

Trunk Performance Report

Trunk Performance Report
Algorithm for
Trunk
Performance
This section includes all the algorithms used in the Trunk Performance
report.
This algorithm estimates, at the busiest intervals in the specified time
period, the number of trunks required to meet your objective blocking
percentage.
6
To determine the actual blocking percentage for a trunk group,
CentreVu CMS uses the following algorithm:
1. Find the trunk occupancy for the trunk group in the average busy
interval for the period of the report.
2. Convert the trunk occupancy into Erlangs (that is, the number of
trunks that are likely to busy at any given point in time). For
example, if at any point in time 8 trunks are likely to be busy, trunk
occupancy is 8 Erlangs. This is the actual carried load for the busy
hour and represents the trunk occupancy due to calls that actually
connected to the switch.
3. Set lower and upper boundaries for offered load. The offered load is
the actual carried load, plus the additional load that would have
occurred if blocked calls had connected to the switch. The initial
upper boundary for the offered load is set at twice the actual carried
load. The initial lower boundary is set at 0.
4. Find the midpoint of the lower and upper boundaries, and use this
midpoint as the offered load in Equation 1.
5. Complete the following equation to find the estimated blocking
percentage P (which will be represented as a decimal between 0
and 1).
n
a
----- -
n !
P = ---------------------------- -
n
+
1
k 1
where a = load in Erlangs (midpoint), n = the number of trunks.
6. Complete Equation 2 to find an estimated carried load. Use the
estimated blocking percentage P found in Step 5 and the offered
load (midpoint) found in Step 4.
carried = offered (1-P)
CentreVu CMS R3V5 Forecast 585-215-825
k
a
-----
k !
6-19
6
(1)
(2)