Markovian ants in a queuing system

Abstract

The synthesis of memoryless Markovian systems and Ant based concept with memory characteristics of deposit pheromone is the basis for the presented artificial intelligence hybrid. Only the initial elements of the system are specified in this paper by illustrating the routes of two ants. The pheromone capacity was first modelled as an exponential-type random variable. The Ant Queueing System was formed. The pheromone capacity was then used to form two independent exponential random variables. The convolution of these variables induces significant quality and quantity changes, mainly the decrease in entropy. The study also provides a possible method for dealing with stationary queueing systems when we are familiar with the state probability and the arrival rate and service rate are unknown. © 2010 Springer-Verlag.