Dynamic Multi Objective Particle Swarm Optimization based on a New Environment Change Detection Strategy

The dynamic of real-world optimization problems raises new challenges to the traditional particle swarm optimization (PSO). Responding to these challenges, the dynamic optimization has received considerable attention over the past decade. This paper introduces a new dynamic multi-objective optimization based particle swarm optimization (Dynamic-MOPSO).The main idea of this paper is to solve such dynamic problem based on a new environment change detection strategy using the advantage of the particle swarm optimization. In this way, our approach has been developed not just to obtain the optimal solution, but also to have a capability to detect the environment changes. Thereby, DynamicMOPSO ensures the balance between the exploration and the exploitation in dynamic research space. Our approach is tested through the most popularized dynamic benchmark’s functions to evaluate its performance as a good method.

💡 Research Summary

The paper addresses the challenge of solving dynamic multi‑objective optimization problems (DMOOPs) where decision variables, objective functions, or constraints change over time. Traditional particle swarm optimization (PSO) and its multi‑objective variant (MOPSO) are well‑suited for static problems but struggle in dynamic environments due to loss of diversity and premature convergence after a change. To overcome these limitations, the authors propose Dynamic‑MOPSO, a novel algorithm that integrates a dedicated environment‑change detection strategy with a reactive re‑initialization mechanism.

The detection component operates at a fixed interval (t = 10 generations). At each interval the algorithm re‑evaluates the current non‑dominated archive and compares the fitness vectors of solutions at time t and t + 1. If any significant degradation is observed, a “change detected” flag is raised. The reaction phase then identifies all particles whose fitness has worsened (negative change) and re‑initializes them, thereby preventing the swarm from being trapped in obsolete local optima. Simultaneously, the archive is refreshed to reflect the new Pareto front. This two‑step process ensures that the swarm continuously balances exploration and exploitation despite the moving landscape.

Dynamic‑MOPSO retains the classic PSO update equations for position and velocity, but introduces stochastic inertia weight (w ∈