Sensitivity examination of the simulation result of discrete event dynamic systems with perturbation analysis

Simulation completed with perturbation analysis provides a new approach for the optimal control of queuing network type systems. The objective of this paper is to calculate the sensitivity range of finite zero-order perturbation, that is, to determine the maximum and minimum size of perturbation wit...

ver descrição completa

Detalhes bibliográficos
Autores: Koltai, Tamas, Larrañeta Astola, Juan, Onieva, Luis, Lozano, Sebastián
Formato: artículo
Fecha de publicación:1994
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2099/4647
Acesso em linha:https://hdl.handle.net/2099/4647
Access Level:acceso abierto
Palavra-chave:Probabilism
Simulation
Perturbation analysis
Queuing networks
Simulació (Matemàtica)
Classificació AMS::65 Numerical analysis::65C Probabilistic methods, simulation and stochastic differential equations
Descrição
Resumo:Simulation completed with perturbation analysis provides a new approach for the optimal control of queuing network type systems. The objective of this paper is to calculate the sensitivity range of finite zero-order perturbation, that is, to determine the maximum and minimum size of perturbation within which zero-order propagation rules can be applied. By the introduction of the concept of virtual queue and first and second level no-input and full-output matrices, an algorithm is provided which can solve this task efficiently in transfer lines and in relatively small general networks when short simulation run is required and the sensitivity of the individual sample path is in question. The implementation of the algorithm with the help of conventional simulation languages is also discussed and presented in an example.