This volume centres upon the important areas of performance evaluation, sensitivity analysis and stochastic optimization of discrete event systems with applications to computer simulation models. It effectively highlights how the score function-likelihood ratio method allows one to evaluate not only the performance, but also to optimize from only a single sample path (simulation) complex discrete event systems, such as queueing networks. A unified and rigorous treatment of the associated stochastic optimization problems is provided and recent advances in perturbation theory encompassed. Throughout the book emphasis is upon concepts rather than mathematical completeness with the advantage that the reader only requires a basic knowledge of probability, statistics and optimization. . This book will be of great interest to students and researchers in operational research, management and industrial engineering, statistics and computer science.