A new simplified MOPSO based on Swarm Elitism and Swarm Memory: MO-ETPSO

TL;DR

This paper introduces MO-ETPSO, a simplified multi-objective PSO algorithm that incorporates swarm memory and elitism, inspired by NSGA-II, demonstrating promising results in vehicle routing problems.

Contribution

The paper proposes a novel simplified MO-ETPSO algorithm integrating swarm memory and elitism, inspired by NSGA-II, enhancing solution retention and convergence.

Findings

Promising results in Green Vehicle Routing Problem

Effective solution retention and convergence

Simplified operators for ease of implementation

Abstract

This paper presents an algorithm based on Particle Swarm Optimization (PSO), adapted for multi-objective optimization problems: the Elitist PSO (MO-ETPSO). The proposed algorithm integrates core strategies from the well-established NSGA-II approach, such as the Crowding Distance Algorithm, while leveraging the advantages of Swarm Intelligence in terms of individual and social cognition. A novel aspect of the algorithm is the introduction of a swarm memory and swarm elitism, which may turn the adoption of NSGA-II strategies in PSO. These features enhance the algorithm's ability to retain and utilize high-quality solutions throughout optimization. Furthermore, all operators within the algorithm are intentionally designed for simplicity, ensuring ease of replication and implementation in various settings. Preliminary comparisons with the NSGA-II algorithm for the Green Vehicle Routing Problem, both in terms of solutions found and convergence, have yielded promising results in favor of MO-ETPSO.