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

7.4.2. Modeling SON for Self-Optimization

From a modeling perspective, the impact of Self-Optimization is going to be structured around the following components:

• 
  Reduced dropped call rate, from DCR_Manual to DCR_SON, which will reduce churn, increase loyalty and therefore improve EBITDA.
  
• 
  Higher sector capacity from a radio perspective, understood as an increase in the amount of traffic that can be served with each carrier in each sector. This gain will reduce/delay the need for new carrier additions. Applying SON techniques, the radio capacity per sector will increase by a factor of G_SON, which is assumed here to be equal for voice and data traffic.
  
• 
  Reduced dispersion of the per sector traffic distribution through load balancing techniques oriented to alleviate Iub congestion problems. Field trials have shown that the standard deviation of the traffic distribution can be reduced by a factor (denoted by X_SON) of around 20% after applying inter-NodeB load balancing techniques for UMTS.
  

7.4.3. Characterizing the Traffic Profile

The exact same models and considerations as in Section 7.3.4 are applied here. The only new element in the model is that, whereas in the Self-Planning analysis the traffic distributions were assumed to be the same before and after the application of SON techniques, in this case the
For every year in the considered time horizon…

Figure 7.16 Summarizing the process to compute the capacity-driven CAPEX and OPEX components of the ROI for Self-Optimization.
utilization of Self-Optimization will reduce the standard deviation of the per sector traffic distribution with a factor denoted by X_SON (see Section 7.4.2), thereby narrowing the distribution.

7.4.4. Modeling the Need for Capacity Expansions

In general, the same methodology that was described in Section 7.3.5 is applied here. In this case, evaluations with and without SON will apply different values for the sector capacity (from a radio perspective) and different traffic distributions, as described in Sections 7.4.2 and 7.4.3. Moreover, two new effects are introduced into the model:

1. 
   Limitations in terms of available spectrum and maximum space for radio equipment in the physical site facility. Although these two effects are unrelated by nature, they are both jointly materialized into a simple limitation in terms of maximum number of carriers per sector, denoted by MAX_carriers. When a sector requires more than MAX_carriers carriers, a new site is assumed to be deployed in its vicinity due to capacity reasons.
   
2. 
   Radio Network Controller (RNC) capacity limitations, characterized by the maximum number of logical cells (RNC_capacity) that can be handled. In this context, the addition of a new carrier in a given sector implies the creation of a new logical UMTS cell. When this limit is exceeded, an additional RNC is installed.
   

The entire process to compute the ROI components due to lower need for capacity expansions is summarized in Figure 7.16.

Customer service
Network
quality (call
performance
Billing and reliability)
12%
32%
Service plan options
14%
14%
Figure 7.17 J.D. Power overall customer satisfaction breakdown [3].