G. Ali, M. S. Memon, A. W. Shaikh


An efficient traffic signals system can reduce the loss of time at great extent, with minimum budget and minimum disturbance to roads. There are many systems widely available, having certain advantages and disadvantages. New approaches  are  emerging  and  old  approaches  are  improving with passage of time. Any  new  approach  cannot  be  practically  applied  at  once  before completion of its cycle and proper testing. Simulation is the most cost effective solution to observe the effectiveness of any proposed system. On the other side Software agents are intelligent entities that work autonomously to achieve their task. The novelty of this paper is integration of Micro-simulation with Multiagent system based on real time approach to add autonomy and elegant behavior through software agents. A simple genetic algorithm and a learning classifier have been integrated in the body of software agent. The user will be able to define realistic models consisting of roads, traffic signals and traffic volume facilitated by suitable and appropriate figures and various property sheets. This type of approach is a unique model where signals are assigned agents of different traffic systems to make a complete traffic management approach. Different approaches will be simulated at once and the agents working at background will generate comparison report of all the simulated approaches. Though  many  simulation  models are available in the market to address various problems but integration of intelligent agents make this model different form the others.

Full Text:



  • There are currently no refbacks.

Copyright (c) 2015 Sindh University Research Journal - SURJ (Science Series)

 Copyright © University of Sindh, Jamshoro. 2017 All Rights Reserved.
Printing and Publication by: Sindh University Press.