M(t)/M(t)/m(t)/C(t) and non-stationary hypercube models

Abstract

Purpose – Verify the effect of the variation of the arrival and service rate during the day in detriment of the equilibrium analysis of the system. Design/methodology/approach – The models M(t)/M(t)/m(t)/C(t) and the non-stationary hypercube were approached considering the change in the number of servers over time. Findings – The study showed how the dynamic approach is more realistic than the equilibrium approach in systems where the variation of parameters is an important factor to be considered. Originality/value – Most studies involving queue systems are based on steady-state analysis of the operation of these systems. However, for certain queuing systems, variations in customer arrival rates, service times and other operating conditions occur within very short time intervals, which makes it difficult to effectively analyze the performance of these systems.