Child Drawing Development Optimization Algorithm based on Child's Cognitive Development

TL;DR

This paper introduces the Child Drawing Development Optimization (CDDO) algorithm inspired by children's cognitive learning stages and the golden ratio, demonstrating superior performance in benchmark optimization problems.

Contribution

The paper presents a novel metaheuristic algorithm based on child cognitive development and the golden ratio, enhancing global search capabilities in optimization tasks.

Findings

CDDO outperforms several state-of-the-art algorithms in benchmark tests.

It effectively avoids local minima and explores promising regions extensively.

The algorithm demonstrates robustness and competitiveness in optimization.

Abstract

This paper proposes a novel metaheuristic Child Drawing Development Optimization (CDDO) algorithm inspired by the child's learning behaviour and cognitive development using the golden ratio to optimize the beauty behind their art. The golden ratio was first introduced by the famous mathematician Fibonacci. The ratio of two consecutive numbers in the Fibonacci sequence is similar, and it is called the golden ratio, which is prevalent in nature, art, architecture, and design. CDDO uses golden ratio and mimics cognitive learning and child's drawing development stages starting from the scribbling stage to the advanced pattern-based stage. Hand pressure width, length and golden ratio of the child's drawing are tuned to attain better results. This helps children with evolving, improving their intelligence and collectively achieving shared goals. CDDO shows superior performance in finding the global optimum solution for the optimization problems tested by 19 benchmark functions. Its results are evaluated against more than one state of art algorithms such as PSO, DE, WOA, GSA, and FEP. The performance of the CDDO is assessed, and the test result shows that CDDO is relatively competitive through scoring 2.8 ranks. This displays that the CDDO is outstandingly robust in exploring a new solution. Also, it reveals the competency of the algorithm to evade local minima as it covers promising regions extensively within the design space and exploits the best solution.