Open AccessMCMC based modelling of queuing systems from empirical data
Author(s) -
Mantas Landauskas,
Eimutis Valakevičius
Publication year - 2011
Publication title -
lietuvos matematikos rinkinys
Language(s) - English
Resource type - Journals
eISSN - 2335-898X
pISSN - 0132-2818
DOI - 10.15388/lmr.2011.mt07
Subject(s) - markov chain monte carlo , computer science , queueing theory , kernel (algebra) , kernel density estimation , construct (python library) , monte carlo method , mathematical optimization , data mining , statistics , artificial intelligence , mathematics , bayesian probability , computer network , estimator , combinatorics
Markov chain Monte Carlo (MCMC) modelling technique requires one to be able to construct a proposal density. There is no universal way to achieve this. This paper considers the universal proposal selection technique based on the kernel density estimate. Two channel queuing system with a priority was modelled using this technique. Empirical data (the observed service times) and the rates of arrival processes are all the informationused for simulating the system.