1. How rapidly will voice traffic be eclipsed by data traffic.
2. How will the rate of the transition impact the final outcome from industry and policy view points.

This research makes original contributions to further the understanding of network traffic trends and their influence on the future telecommunications policy and business environments. Novel aspects of this work are:

· The development of a model to characterize the transition of telecommunications traffic mix. This model is extended in the Thesis to characterize the transition of telecommunications revenue.

· A calibration of the change in network voice and data traffic composition. A parallel calibration is also performed (only in the Thesis) for the accompanying change in network voice and data revenues.

· The observation that a lag exists between the traffic and revenue crossover events.

· The suggestion that the rapidity of the transition, as opposed to the change itself, is a primary determinant of the post "crossover" telecommunication industry structure.

· Specific observations concerning the consequences of this sudden shift with respect to IP voice substitution effects, market economics, and Universal Service.

Packet based, e.g., Internet, traffic will soon surpass conventional forms of wired communication, e.g., circuit-switched voice traffic. Moreover, the growth in packet traffic is so pronounced that by 2007, telephony may constitute only 10% of overall backbone traffic. In this work, we investigate this transition of the communication network from a "voice-centric" to a "data-centric" model. In particular, we examine the speed with which data traffic will eclipse voice traffic, and the significance of the shift’s suddenness to the final outcome. While a prediction concerning the precise timing of the circuit-to-packet crossover is of some interest, this report’s focus is on addressing the fundamental issues relevant to the rapid evolution of a data dominated communication network. To aid in understanding the shift, we present and characterize a traffic transition model. Based on this model, we determine that the window in which data traffic increases to contribute 90% of overall backbone traffic is 10 years from today, and find that the crossover (50%) point occurs in September 1998. We argue that the suddenness of the change, as opposed to the change itself, will be a key issue in determining the post "crossover" telecommunication industry structure, and discuss the implications of this finding for ISPs, RBOCs, switch equipment vendors, and Universal Access. In the thesis this model is extended to investigate the revenue crossover that accompanies the traffic crossover and the relationship between the two events is characterized.