Telektronikk 2/3.2001
technical performance requirements, and so
forth. Economic aspects may include require-
ments on net present value, cash flow restriction,
financial conditions, etc. Figure 2 contains an
illustration similar to the one in Figure 1, with
more emphasis placed on the economic side, on
cost and revenue in particular. In order to arrive
at a tractable model, quite a few approximations
are made on the technical side.
A techno-economic study will capture a number
of time periods (e.g. years). The variables must
then refer to a number of the time periods, or an
evolution of the variables must be given. Two
examples of traffic load are given in Figure 3.
Starting with traffic load for individual applica-
tions, the total traffic load expected from a user
can be estimated by looking at the simultaneous
use of the applications. Commonly a set of refer-
Figure 2 Illustration of work flow for techno-economic studies
Figure 3 Examples of traffic load inputs for a 10 year study period (left – aggregate traffic load, right – fraction of the traffic load
referring to customer segments)
Access
solutions
Economic Data
Traffic
load
Capacities and roll out plan
Calculation
Market
Architecture
Demands modelling
Applications
demands Network topology
Applications
characteristics Operationalcosts
Network
elements
Tariffs
Market
shares
Traffic
flow
Regulations
NPV, IRR
Investments
Running costs
Revenues
Cash flows
Depreciations
Residential
Corporate
Large SMB
Small SMB
Micro SMB
12 345 67 8910
100
%
80
60
40
20
0
Time period (year) Time period (year)
12345678910
450
400
350
300
250
200
150
100
50
0
Traffic load (Gb/s)
Irena Grgic (30) is Research
Scientist at Telenor R&D, Kjeller.
She is mainly involved in activi-
ties related to QoS and charging
for different networks and sys-
tems, and studies related to net-
work evolution, both in interna-
tional and national projects. She
holds an MSc in Electrical Engi-
neering from the University of
Zagreb in 1999.
[email protected]
88
ence periods is assumed for these calculations
(e.g. similar to some “busy hour” although a
shorter period than one hour would likely be
assumed). The reference periods may refer to
morning, afternoon, evening, or any other practi-
cal identification.
Given these loads, the performance targets and
the capacities of the network elements, a roll-out
plan for the network arises. An aggregated plan
showing only the number of elements is given in
Figure 4. Then, having the number of units
needed for each year and the cost for each of the
units, a cash flow can be obtained. Here the
demands may also be related to a set of revenue
streams where a mixture of subscription rate,
usage rate and income from other parties (e.g.
for commercial) may differ for the different
applications and user groups.
