The design of forward looking cost models for local exchange telecommunications networks

Abstract

In the last few years great strides have been made in the development of algorithms that design telephone networks. As computational capabilities improve it is possible to produce better results, both from an engineering and an economic standpoint. This paper considers the design issues that cost model developers have addressed successfully. Many of these issues are illustrated by a detailed description of a model developed by FCC staff, known as the Hybrid Cost Proxy Model (HCPM). HCPM is capable of utilizing very precise customer location data. From these data, the model uses clustering algorithms to identify serving areas that satisfy appropriate engineering constraints. Within each serving area, the model uses a modified minimum-cost spanning tree algorithm to connect actual customer locations to a serving area interface. The same tree algorithm connects each interface point to a switch. Within each path, the model performs intensive integer searches to find the cost minimizing, yet engineering-feasible, choice of technology and electronics for that path. The result is a-low cost, feasible network plan that gives an appropriate estimate of the forward looking cost of providing wireline telephone service to a particular area. This estimate should prove particularly useful in the ongoing debate about the size and make-up of the Universal Service Fund, and for other regulatory purposes such as the pricing of interconnection and unbundled network elements.