#TITLE_ALTERNATIVE#
Traffic demand always varies with the time. It makes traffic forecasting to be one of the key factors in business communication. The estimation about the changing <br /> <br /> <br /> in traffic demand especially the increasing in traffic at service networking is causing an accre...
Saved in:
Main Author: | |
---|---|
Format: | Final Project |
Language: | Indonesia |
Online Access: | https://digilib.itb.ac.id/gdl/view/14905 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Institut Teknologi Bandung |
Language: | Indonesia |
Summary: | Traffic demand always varies with the time. It makes traffic forecasting to be one of the key factors in business communication. The estimation about the changing <br />
<br />
<br />
in traffic demand especially the increasing in traffic at service networking is causing an accretion in network capacity which is also related to the quality in service. <br />
<br />
<br />
In this final project, I am using traffic forecasting in cellular communication with the time series model which is Autoregressive Integrated Moving Average (ARIMA) by using Statistical Package for the Social Sciences (SPSS) application and Eviews. The experiment performed only on a certain condition of time such as Idul Fitri, Christmas – New Year. The result of this experiment shows that ARIMA model is compatible to be used in traffic demand forecasting in cellular communication, but the forecasting <br />
<br />
<br />
should be done in every BSC one by one. The reason of doing so is because they have unique characteristic. The correction factor should be found specifically for feast <br />
<br />
<br />
day. The application of every model is also followed by the calculation of model’s. In BSC_04 the appropriate model is ARIMA(1,0,1) with 4.51% MAPE and on Idul Fitri is added by correction factor 1.8049/1.7222. In BSC_08 the appropriate <br />
<br />
<br />
model is ARIMA(1,0,1) with 5.08% MAPE and on Idul Fitri is added by correction factor 1.1554/1.1239. In BSC_10 the appropriate model is ARIMA(1,0,1) with 3.76% MAPE and on Idul Fitri is added by correction factor 0.9453/0.9671. In BSC_09 the appropriate model is ARIMA(0,0,1) with 3.56% MAPE and on Idul Fitri is added correction factor 0.9602/1.2196. ARIMA model is hoped by every telecommunication’s operator to provide a better network in terms of capacity, quality, and signal coverage so that can give satisfaction to their customer. Moreover there will be not any waste of resources. |
---|