Gaussian Transfer Functions Based Binary Particle Swarm Optimization for Enhanced Performance in Un-capacitated Facility Location Problem
Year: 2024
Type: article
Abstract: This study introduces Gaussian Binary Particle Swarm Optimization (G-BPSO), designed to address binary optimization challenges effectively. G-BPSO employs new transfer functions of the Gaussian type derived from the power functions to enable mapping of real-valued vectors of individual encodings into binary form. This ensures smooth change between steps and improved convergence. To assess the effectiveness of G-BPSO, a host of complex optimization problems such as the un-capacitated facility location problem are investigated. Enhanced efficiency and improvement over existing methods in binary optimization is observed. The MATLAB code of G-BPSO is made open-access through https://github.com/kanak02/GBPSO.
Source: MM Science Journal
Authors: Kanak Kalita, Lenka Cepova, Pradeep Jangir
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4405003023