Optimizing Telecom Fiber Network Deployment Using Neural Networks: A Predictive Approach to Enhancing Customer Reach

Abstract

In the ever-developing industry of telecommunication, strategic positioning of the fiber optic network has to be strategic to ensure it has penetrated into its target market to the max level with little or minimum investment. This paper develops a strategy of fiber placement of telecom aimed to cover the maximum number of new clients using a neural network approach. infusion of various geographic, demographic and economic factors in a model that can predict areas that will require optimal fiber deployment. Such variables as population density, level of urbanization, and the closeness of the major cities are constructed from related data sources. From actual data, the model presented in this paper is trained and validated, suggesting that its application could greatly improve the process of expanding telecom systems. This paper also presents information on the structure of the neural network, the implementation of plans and part of the performance test. The future work includes exploring other approaches of machine learning
and discussing the threats connected to the large scale use of the network.