Thedy, JohnJohnThedyLiao, Kuo-WeiKuo-WeiLiaoWibowo, Mochamad AgungMochamad AgungWibowo2026-04-242026-04-242026-08https://www.sciencedirect.com/science/article/pii/S2772375526003175https://scholars.lib.ntu.edu.tw/handle/123456789/737575This study aims to develop a smart structural optimization framework for automated greenhouse design to support data-driven decision-making in agricultural production environments. To address the continued reliance on empirical and trial-and-error practices under complex, geometry-dependent loading conditions, the proposed framework integrates parametric design automation with load-responsive optimization, enabling dynamic coupling between structural configuration and external loads. The methodology combines an ABAQUS-based parametric modeling system with a weighted multi-objective symbiotic organism search (WMOSOS) algorithm to simultaneously minimize material consumption and maximum von Mises stress ratio, generating a Pareto front of non-dominated solutions. The framework is validated through standard ZDT and 25-bar truss benchmark problems, demonstrating improved convergence performance and solution diversity compared with established multi-objective optimization algorithms. A full-scale greenhouse case study further illustrates the applicability of the proposed approach in producing multiple feasible design alternatives that balance structural performance and cost. The main contribution of this study lies in the integration of automated structural design, geometry-dependent load updating, and multi-objective optimization into a unified framework tailored for smart greenhouse systems. To promote transparency and reproducibility, the ABAQUS-based parametric modeling scripts and the WMOSOS algorithm are made publicly available.GreenhouseOptimizationAutomationAgriculture, Metaheuristicjournal article10.1016/j.atech.2026.102096