A law of the iterated logarithm for extreme queue length in multiphase queues (Q2385539)

In this paper the research on the law of the iterated logarithm (LIL) under the conditions of heavy traffic in multiphase queueing systems (MQS) has been considered. We recall that in the MQS the customer does not visit the same queueing node twice. Such systems are treated as a special case of the open Jackson network. The author has proved some theorems on the LIL for a cumulative process associated with the queue length of customers and the waiting time of the customer, the virtual waiting time of the customer in an ordinary queueing system GI/G/1. The main results consist of theorems on the LIL for the extreme values of the main probability characteristics of MQS in heavy traffic (maxima and minima of the queue length of customers). The theorems presented in the paper are proved for a class of MQS with the service discipline FCFS and the endless waiting time of customers in each phase of the queue, when the time portions between the customers arriving at the MQS are independent identically distributed random variables.