A novel self-organisation model for improving the performance of permutation coded genetic algorithm Online publication date: Fri, 11-Sep-2020
by K. Dinesh; J. Amudhavel; R. Rajakumar; P. Dhavachelvan; R. Subramanian
International Journal of Advanced Intelligence Paradigms (IJAIP), Vol. 17, No. 3/4, 2020
Abstract: Genetic algorithms (GA) are extremely powerful among evolutionary principles and being used in variety of fields for solving more complex problems. Varieties of assistive techniques have been proposed to improve the performance of genetic algorithms w.r.t. the nature of the application and self-organisation is one such model, which is aimed at improving the performance of the GAs by all means. The self-organisation models enable the systems to acquire and maintain the structure by themselves, without any external control. It is highly evidenced that it gives greater benefits to solve the complex problems with competent efficiency levels in conjunction with the classical GAs. The combined version of SOM and GA has the power of better exploration. In this way, the work reported in this paper proposes an efficient pattern based self-organisation model for improving the performance of the GA for the combinatorial optimisation problem. The competency of the proposed model is demonstrated by means of a set of well-defined experiments over the selected benchmark travelling salesman problem (TSP) instances. The assessments proved the efficiency of the technique in terms of a set of generic performance criteria like convergence rate, convergence time, error rate, nearest neighbour ratio and distinct individuals.
Online publication date: Fri, 11-Sep-2020
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Advanced Intelligence Paradigms (IJAIP):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email firstname.lastname@example.org