Analysis of bayesian method in queuing system customer service locket Semarang Tawang railway station

Abstract

Jamming is one of the serious problems in Indonesia caused by the increase of vehicles. The government has been made solution for this situation for example was public transportation. The train is one of the suitable public transportation because of the ticket price was cheap. Tawang Railway Station Semarang was the biggest railway station in Semarang. In the specific day such long holiday or celebration day, many people have chosen train to take them. This make a queuing situation on the counter of station especially on customer service locket which provide information about the station and more. Queue theory models provide the random of arrival and service time. The Bayesian theory suits to handle the problem of queuing that has been working for several times to repair the prior information with new information of sample and also can make a combination of the distribution after assumption steady state. Based on the analysis of the queue models for customer service are (G/G/c):(GD/∞/∞) from the posterior distribution with combination from prior distribution and likelihood sample. The prior distribution used in this research is Poisson. The likelihood sample used Uniform Discrete and Negative Binomial distribution. The posterior distribution is a combination of Uniform Discrete and Beta distribution. Queue models can be used to count the size of the system performance. Based on the calculations and analysis with RGui, it can be concluded that the queuing system of customer service locket have been good because its steady state and busy probability is higher than jobless probability.