System and method for predicting key performance indicator (KPI) in a telecommunication network

A system for predicting Key Performance Indicator (KPI) in a telecommunication network, the system comprising:

• a memory;
• a processor coupled to the memory, wherein the processor is configured to execute programmed instructions stored in the memory for:
• monitoring a set of counters and a KPI, corresponding to a telecommunication network, wherein the set of counters and the KPI are monitored for a predefined time interval to gather sample data;
• analyzing the sample data using a data analysis technique to:
• identify a subset of counters, from the set of counters, influencing the KPI, and
• identify a correlation coefficient associated with each counter from the subset of counters, wherein the correlation coefficient associated with each counter is identified after normalizing the subset of counters;
• and
• applying regression models on the subset of counters and the KPI in order to build a correlation equation between the subset of counters and the KPI, wherein a regression model is selected based on how responsive a prediction has to be which varies depending on a KPI a user wants to model,
• forecasting and predicting performance of a KPI using the correlation equation.

A non-transitory computer readable medium embodying a program executable in a computing device for predicting Key Performance Indicator (KPI) in a telecommunication network, the computing device executing the program comprising:

• a program code for monitoring a set of counters and a Key Performance Indicator, corresponding to a telecommunication network, wherein the set of counters and the KPI are monitored for a predefined time interval to gather sample data;
• a program code for analyzing the sample data using a data analysis technique to:
• identify a subset of counters, from the set of counters, influencing the KPI, and identify a correlation coefficient associated with each counter from the subset of counters, wherein the correlation coefficient associated with each counter is identified after normalizing the subset of counters;
• and
• a program code for applying regression models on the subset of counters and the KPI in order to build a correlation equation between the subset of counters and the KPI, wherein a regression model is selected based on how responsive a prediction has to be which varies depending on a KPI a user wants to model,
• a program code for forecasting and predicting performance of a KPI using the correlation equation.