# System Identification Theory for Queuing Networks

**Project level:** PhD, Masters

Queueing networks are stochastic mathematical models that are often used for analyzing service, manufacturing and communication systems. One often attempts to model the situation at hand by means of a queueing network and then uses the resulting model to gain insight into system performance metrics. In cases where the appropriate model is well identified and the parameters are well estimated, this methodology often proves useful. But in other cases, estimating the parameters and identifying the model is a real challenge. Most of the theoretical research of queueing networks has focused on the properties of the resulting stochastic processes, assuming that the queueing network model and its parameters are known. The model fitting and parameter aspects are often ignored. This leaves practitioners with the task of fitting models and estimating parameters in an ad-hoc manner. The purpose of this project is to devise, analyze and explore model fitting and parameter estimation methods for queueing networks. The results may prove useful in applications of health-care logistic systems and other business processes.