Sensitivity analysis and cost optimization for state-dependent M/G/1/K queue with retrial orbit and admission control F -policy
Sudeep Singh Sanga,Darshan Rameshbhai Dhapa,Nidhi
DOI: https://doi.org/10.3934/jimo.2024183
2024-12-09
Journal of Industrial and Management Optimization
Abstract:This study investigates the state-dependent M/G/1/K queueing model, which includes realistic features such as retrial orbit and admission control F-policy. When the server is busy, arriving customers are compelled to join the retrial orbit and wait until the server becomes available to serve them. Moreover, the admission control F-policy stipulates that the system cannot accommodate more customers once it reaches its maximum capacity (say ) until the system size attains its predefined threshold value F. The Chapman-Kolmogorov (C-K) equations are formulated using the supplementary variable technique (SVT) to conduct a mathematical analysis. Subsequently, the C-K equations are solved using the Laplace-Stieltjes transform (LST), followed by the recursive method. We deduce the machine repair problem (MRP) and time-sharing model (TSM) from the general state-dependent M/G/1/K queueing model. Moreover, several system performance measures are also developed, analyzing the state-dependent model. We provide a numerical example to demonstrate the effect of the input queueing parameters on system performance measurements in both cases, i.e., MRP and TSM. Additionally, we formulate the cost function with decision variables such as the threshold parameter and service rate. To acquire the minimum cost of the system, we utilize the direct search method (DSM) and the particle swarm optimization (PSO) algorithm. To exhibit the applicability of the state-dependent model in real-world situations, we showcase the application of both MRP and TSM.
engineering, multidisciplinary,operations research & management science,mathematics, interdisciplinary applications