Title: Investigating the population dynamics of differential evolution algorithm for solving multi-objective RFID reader placement problem

Authors: K. Devika; G. Jeyakumar

Addresses: Department of Electrical and Electronic Engineering, Amrita School of Engineering, Coimbatore, Amrita Vishwa Vidyapeetham, India ' Department of Computer Science and Engineering, Amrita School of Engineering, Coimbatore, Amrita Vishwa Vidyapeetham, India

Abstract: Evolutionary algorithms (EAs), are the bio-inspired algorithms to solve optimisation problems. Numerous works are in the pipeline for studying their behaviour on complex problems. This work is carried out in three phases. In phase 1, a comparative analysis of two population initialisation (PI) techniques of EAs was done. The results showed that DE performs better with opposition-based learning PI (OBLPI) than with the random PI (RPI) technique. The phase 2 analysed the performance of DE in solving the multi-objective optimisation problems (MOOP) with RPI and OBLPI techniques. This analysis revealed that DE with OBLPI and RPI performed well for problems with lower and higher population sizes, respectively. In phase 3, DE was used solving the radio frequency identification (RFID) reader placement problem, for buildings with single room and multiple rooms of different sizes. The design of experiment, the results obtained and the inferences are presented in this paper.

Keywords: evolutionary algorithms; differential evolution; population initialisation; multi-objective optimisation problem; MOOP; RFID reader placement problem; variable length chromosomes; VLC; fixed length chromosomes; FLC; opposition-based learning PI; OBLPI.

DOI: 10.1504/IJISTA.2021.119027

International Journal of Intelligent Systems Technologies and Applications, 2021 Vol.20 No.2, pp.82 - 102

Received: 23 May 2020
Accepted: 06 Oct 2020

Published online: 12 Nov 2021 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article