Forecasting Techniques for Managing Mobile Data Network Capacity and Growth

This paper proposes a two part methodology and modeling approach, covering techniques for managing mobile data network capacity and growth using Bell Labs forecasting and predictive techniques. The combined methodology overcomes the shortcomings of traditional forecasting techniques by using actual traffic data from the service provider's network and the operator's "run-the-business" groups to obtain a detailed user behavior model. Subscriber forecast input from their marketing organization, along with the behavioral trends and Bell Labs forecasting algorithms, methodologies and tools allow for the generation of both near term and longer term forecasts.