Return on Investment (roi) for Multi-Technology son juan Ramiro, Mark Austin and Khalid Hamied

Figure 7.12 Overall number of carriers that are required to serve all the offered traffic.
Table 7.1 Assumptions for the ROI assessment of Self-Planning

Network size and initial configuration
Number of sites	10000
Sectors per site	3
Initial number of carriers per sector (year 1)	2
Initial number of E1 links per site (year 1)	2

Manual process accuracy (conservative error rates, in line with field observations)

Over-dimensioning rate (o)	20%
Under-dimensioning rate (u) Initial average traffic per sector in the busy hour (year 1)	2%
Voice (R99)	5 Erlangs
Data (HSDPA) Traffic growth	300 kbps
Relative traffic growth pattern (%; years 2-10) Cost structure	[10, 10, 10, 10, 10, 10, 10, 10, 10] %
Annual OPEX per E1link	€5000
Man-days to dimension 1000 sites	40
OPEX per engineer and day	€300
OPEX to operate an ideal SON solution	€0
CAPEX for extra carrier per sector Capacity figures and other radio indicators	€15000
Radio interface capacity for voice only (R99)	21 Erlangs
Radio interface capacity for data only (HSDPA)	1600 kbps
Nett Iub capacity for voice only (R99, per E1 link)	25 Erlangs
Nett Iub capacity for data only (HSDPA, per E1 link)	630 kbps
Soft Handover Overhead	40%

1400.0
1200.0
1000.0
800.0
600.0
400.0
200.0
0.0
1 2 3 4 5 6 7 8 9 10
Year
Figure 7.13 Ideal calculation (carried out by the Self-Planning function) of the required carrier expansions (per year) to cope with the experienced traffic growth.

1400.0
1200.0
1000.0
800.0
600.0
400.0 200.0
0.0
1 2 3 4 5 6 7 8 9 10
Year
Figure 7.14 Ideal versus manual calculation of the required carrier expansions (per year) to cope with the experienced traffic growth.
offered traffic is illustrated in Figure 7.13. As can be seen, this number equals 166 for year 1, and then evolves as a function of the annual traffic growth. The need for carrier additions in the first year is larger than in the second year because of the initial configuration of the network, which is under-equipped. Since traffic is assumed to grow 10% each year, the subsequent need for new carriers grows exponentially, due to the fact that annual absolute traffic increases become higher every year.
For the manual case, the amount of extra carrier additions that imply the actual purchase of additional equipment is illustrated in Figure 7.14. This calculation is conducted according to the assumptions presented in Section 7.3.3 about the manual dimensioning process regarding the handling of capacity surplus and dimensioning errors carry-over from year to year.
Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10
CAPEX reduction 0.45 –0.14 0.16 0.22 0.28 0.35 0.43 0.51 0.59 0.67
Extra carriers 0.45 –0.14 0.16 0.22 0.28 0.35 0.43 0.51 0.59 0.67
OPEX reduction 8.32 1.66 1.86 2.08 2.32 2.57 2.84 3.13 3.42 3.72
E1 expansions 8.20 1.54 1.74 1.96 2.20 2.45 2.72 3.01 3.30 3.60
Automation 0.12 0.12 0.12 0.12 0.12 0.12 0.12 0.12 0.12 0.12
Total (€ million) 8.77 1.52 2.02 2.30 2.60 2.92 3.27 3.64 4.01 4.39

Figure 7.15

Cash flow projection for Self-Planning.

Note that, to facilitate the comparison, Figure 7.14 also reproduces the number of carrier additions in the case in which the capacity planning process is error-free due to the application of advanced Self-Planning techniques.
As can be seen in Figure 7.14, the difference between ideal and manual calculations is not as big as the 20% over-estimation rate assumed in Table 7.1, due to the carry-over and compensation mechanisms that have been assumed for over-dimensioned items (see Section 7.3.3), which basically imply that the unnecessary carriers that are added during one year can be flexibly used (through cost-free HW relocation) during subsequent years, thereby decreasing the effective amount of purchases that need to be carried out. Also due to the application of these carry-over mechanisms, during the second year, the need for additional carriers with the manual methodology is lower than with Self-Planning, which is translated into a negative contribution of the CAPEX-related differential cash flow component of that year (this is illustrated in Figure 7.15). Note that this effect is present every year. However, in the second year the combination of a big need for investment during the first year (due to the original network underdimensioning) followed by a modest traffic increase in the second year has lead to such extreme values that the value of the differential cash flow related to CAPEX becomes negative. In other words, the manual process over-invested heavily during the first year and then made use of that investment in the second year. As stated before, this means that the application of Self-Planning techniques delays the need for investment, which has a positive impact on the NPV, even though some individual cash flow components are negative (see Figure 7.3).
Once the amount of HW to be purchased has been calculated, it is possible to compute the corresponding cash flow components by simply multiplying the amount of required carriers per year with the assumed monetary cost per carrier.
Going forward, a very similar illustration can be done for the addition of E1 links, which are considered as annual recurrent OPEX in this exercise. Since they are very similar processes, details are not provided here for the sake of brevity. Applying the described methodology, the cash flow projection presented in Figure 7.15 can be built and, assuming a discount rate of 20%, the resulting NPV (described in Equation 7.1) is €15.92 million.