Estimation and Prediction in Multi-server Markovian Queueing System with Reverse Balking
摘要
This article investigated classical and Bayesian inference methods for estimating the traffic intensity parameter within a novel multiserver Markovian queueing model that incorporates reverse balking. We have provided a comparative analysis of both the approaches. The findings demonstrate that Bayesian estimates perform superior than classical maximum likelihood (ML) estimates in terms of root mean square errors (RMSE). Additionally, we computed predictive probabilities of the number of customers in the system under various hyper-parameter values. The results provide valuable insights for efficient queue management and improving service efficiency in systems where reverse balking occurs. Additionally, real-life applications of the proposed queueing model are explored to enhance the understanding of the methodology employed in the study.