No abstract

The increasing demand by customers and the exponential growth in the telecommunication services in Ghana require a reliable telecommunication networks. To improve on the satisfaction of customers, it is incumbent on telecommunication network operators to ensure network availability. Designing resilient networks with low downtime and low recovery time is of paramount importance. This study describes a data set of daily outage measurement collected at the Network Operation Centre (NOC) from January 2012 to December 2012. A statistical analysis of the data was done to model telecommunication network availability pattern. Following the Box-Jenkins approach, Autoregressive Integrated Moving Average (ARIMA) was employed to analyze daily network outage. The study was mainly intended to forecast the telecommunication network outage duration in Ghana. The steps of model identification, parameter estimation, and diagnostic checking are performed as recommended. A model-selection strategy based on the corrected Akaike Information Criterion (AICc), Bayesian Information Criterion and Akaike Information Criterion (AIC) are adopted to determine the correct model specification. The results of the study indicate that ARIMA (2, 0, 2) model is suitable in predicting the network outage of telecommunication networks. Using five statistical indicators, namely, Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), Mean Square Error (MSE), Mean Percentage Error (MPE) and Mean Absolute Error (MAE), the prediction accuracy of the proposed model is compared with the accuracies obtained with other results. The results showed that the outages and recovery time of a telecommunication network can be predicted which will help in fault management, network planning and optimization.

No abstract

No abstract