Dynamic queueing models for the operations management of communication services

The communications industry is spawning innovative services and systems in the post-Internet age. The industry is growing rapidly due to reduced equipment costs, intense competition and deregulation. Effective management of these new services is often difficult due to uncertainty and fluctuating demand. The service manager would like to maximize profit while maintaining a good quality of service for the customer experience. In this dissertation, four fundamental operation management problems for communication services are identified and addressed: performance modelling as well as provisioning, pricing, and priority decision making.; Previous attempts to solve these operational problems rely on steady state or stationary results. However, the steady state assumption is not valid for many communication services. We combine dynamic queueing models with dynamic optimization to tackle these four management problems. Random measures, strong approximation theory and the calculus of variations are the foundational theoretical tools used to build, analyze, and optimize our models.