Evolutionary Algorithm-Based Design and Performance Evaluation of Wood–Plastic Composite Roof Panels for Low-Cost Housing

Abstract

Wood–plastic composites (WPCs) have emerged as a sustainable and cost-effective material for construction, particularly in low-cost housing solutions. However, designing WPC panels that meet structural, serviceability, and manufacturing constraints remains a challenge. This study focused on optimizing the cross-sectional shape of WPC roof panels using evolutionary algorithms to minimize material usage while ensuring compliance with deflection and stress constraints. Two evolutionary algorithms—the genetic algorithm (GA) and particle swarm optimization (PSO)—were employed to optimize sinusoidal and trapezoidal panel profiles. The optimization framework integrated finite element analysis (FEA) to evaluate structural performance under uniformly distributed loads and self-weight. The modulus of elasticity of the WPC material was determined experimentally through three-point bending tests, ensuring accurate material representation in the simulations. The trapezoidal profile proved to be the most optimal, exhibiting superior deflection performance compared with the sinusoidal profile. A comparative analysis of GA and PSO revealed that PSO outperformed GA in both solution optimality and convergence speed, demonstrating its superior efficiency in navigating the design space and identifying high-performance solutions. The findings highlight the potential of WPCs in low-cost housing applications and offer insights into the selection of optimization algorithms for similar engineering design problems.