Firefly Algorithm Applications in Optimization Problems

The firefly algorithm is a bio-inspired metaheuristic grounded in the flashing behaviour of fireflies. In this model, candidate solutions are treated as fireflies whose brightness represents objective-function values. Less bright individuals move towards more luminous ones according to an attractiveness function that decays with distance, while a randomisation term ensures diversity. Since its introduction, the firefly algorithm has been adapted and hybridised to address a wide spectrum of optimisation challenges, including continuous and combinatorial problems, engineering design, image processing, parameter estimation in biological models, and control-system tuning. Key developments have focused on balancing global exploration and local exploitation, accelerating convergence, avoiding premature stagnation, and automating parameter selection. Practical applications range from structural design and energy management to semantic segmentation and network routing, underlining the global significance of the approach in delivering robust and computationally efficient solutions to complex, multimodal landscapes.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

A recent work in semantic segmentation has employed an enhanced firefly algorithm to optimise hyperparameters of deep convolutional networks. By integrating mutation operators, a neighbourhood search strategy and adaptive sine–cosine trajectories, the authors generated multiple fine-tuned variants of a benchmark segmentation model. An evolving ensemble of these variants demonstrated state-of-the-art performance on underwater and medical imaging datasets, outperforming both standard deep-learning architectures and classical metaheuristics.

In engineering optimisation, a novel variant termed FA1→3 introduced three distinct movement strategies to improve global exploration and convergence characteristics. Evaluated on a standard suite of test functions, this method exhibited superior robustness and solution quality compared with the original firefly algorithm and several advanced alternatives. Further application to six real-world design problems confirmed its efficacy in locating optimal design variables with higher accuracy and reduced computational effort.

Another contribution presented a fast firefly algorithm designed for function optimisation and control of a brushless DC motor. By accelerating the convergence of brightness-based movements and retaining exploration capacity, this variant achieved rapid attainment of optimal parameters for a Proportional–Integral regulator. Comparative studies against particle swarm optimisation and genetic algorithms demonstrated its superior convergence speed while maintaining comparable precision.

Firefly Algorithm Applications in Optimization Problems publication trend

The graph below shows the total number of articles in firefly algorithm applications in optimization problems across all publications each year (not limited to Nature Index journals).

Technical terms

Metaheuristic algorithm: A high-level framework guiding subordinate heuristics to explore solution spaces for complex optimisation problems.

Brightness: A measure of solution quality that influences the movement of fireflies towards promising regions.

Attractiveness: A function of brightness and distance that governs the probability and extent of attraction between fireflies.

Exploration vs Exploitation: Exploration refers to global search for diverse solutions, whereas exploitation denotes intensive local search around the best solutions.

Convergence rate: A metric indicating the speed at which an algorithm approaches an optimal or satisfactory solution.

Hyperparameter: A configuration setting that defines algorithmic behaviour and must be tuned for optimal performance.

References

1. Semantic segmentation using Firefly Algorithm-based evolving ensemble deep neural networks. Knowledge-Based Systems (2023).
2. A new firefly algorithm with improved global exploration and convergence with application to engineering optimization. Decision Analytics Journal (2022).
3. A Fast Firefly Algorithm for Function Optimization: Application to the Control of BLDC Motor. Sensors (2021).