Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ABCO: Adaptive Bacterial Colony Optimisation

Barisi Kogam, Yevgeniya Kovalchuk, Mohamed Medhat Gaber

TL;DR

Optimization seeks efficient, accurate solutions to find the global optimum of $f(x)$. The authors propose Adaptive Bacterial Colony Optimisation (ABCO), an E. coli foraging-inspired method that integrates three stages—explore, exploit, reproduce—along with neighbor-informed movement and weighted-average reproduction to adapt the search. Empirical comparisons against PSO and ACO on ten benchmark functions show ABCO can achieve competitive accuracy with reduced runtime, particularly at smaller population sizes, demonstrating the value of an adaptive explore–exploit balance. The work highlights the practical potential of locally guided, adaptive search strategies for fast convergence in continuous optimisation and suggests avenues for real-world evaluation and alternative reproduction schemes.

Abstract

This paper introduces a new optimisation algorithm, called Adaptive Bacterial Colony Optimisation (ABCO), modelled after the foraging behaviour of E. coli bacteria. The algorithm follows three stages--explore, exploit and reproduce--and is adaptable to meet the requirements of its applications. The performance of the proposed ABCO algorithm is compared to that of established optimisation algorithms--particle swarm optimisation (PSO) and ant colony optimisation (ACO)--on a set of benchmark functions. Experimental results demonstrate the benefits of the adaptive nature of the proposed algorithm: ABCO runs much faster than PSO and ACO while producing competitive results and outperforms PSO and ACO in a scenario where the running time is not crucial.

ABCO: Adaptive Bacterial Colony Optimisation

TL;DR

Optimization seeks efficient, accurate solutions to find the global optimum of . The authors propose Adaptive Bacterial Colony Optimisation (ABCO), an E. coli foraging-inspired method that integrates three stages—explore, exploit, reproduce—along with neighbor-informed movement and weighted-average reproduction to adapt the search. Empirical comparisons against PSO and ACO on ten benchmark functions show ABCO can achieve competitive accuracy with reduced runtime, particularly at smaller population sizes, demonstrating the value of an adaptive explore–exploit balance. The work highlights the practical potential of locally guided, adaptive search strategies for fast convergence in continuous optimisation and suggests avenues for real-world evaluation and alternative reproduction schemes.

Abstract

This paper introduces a new optimisation algorithm, called Adaptive Bacterial Colony Optimisation (ABCO), modelled after the foraging behaviour of E. coli bacteria. The algorithm follows three stages--explore, exploit and reproduce--and is adaptable to meet the requirements of its applications. The performance of the proposed ABCO algorithm is compared to that of established optimisation algorithms--particle swarm optimisation (PSO) and ant colony optimisation (ACO)--on a set of benchmark functions. Experimental results demonstrate the benefits of the adaptive nature of the proposed algorithm: ABCO runs much faster than PSO and ACO while producing competitive results and outperforms PSO and ACO in a scenario where the running time is not crucial.
Paper Structure (6 sections, 4 figures, 9 tables, 3 algorithms)

This paper contains 6 sections, 4 figures, 9 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Related work
  3. Proposed ABCO algorithm
  4. Experiments
  5. Results and Discussion
  6. Conclusion

Figures (4)

  • Figure 1: The proposed ABCO algorithm illustration: (a) exploration stage; (b) exploitation stage; (c) reproduction stage.
  • Figure 2: Experiment 1 (population size of 100): error rate and runtime results of the ACO, PSO and ABCO algorithms over 50 runs for each of the ten test functions (error rates are shown in the first and third columns, while runtime -- in the second and fourth columns).
  • Figure 3: Experiment 2 (population size of 25): error rate and runtime results of the ACO, PSO and ABCO algorithms over 50 runs for each of the ten test functions.
  • Figure 4: Experiment 3: error rate and runtime results of the ACO, PSO and ABCO algorithms over 50 runs for each of the ten test functions. ABCO population size set to 25 (15 for the Sphere function); ACO and PSO population sizes set to 100.