Plant production yield optimization and cost-effectiveness using an innovative artificial multiple intelligence system

The endeavor to augment the productivity of agricultural commodities is fundamentally contingent upon the establishment of optimal cultivation conditions, particularly in the realm of hydroponics greenhouses. Here, the failure to meticulously regulate environmental parameters may precipitate escalated production expenditures. Our research is primarily focused on the optimization of plant growth parameters, with the objective of maximizing production yield whilst concurrently minimizing associated costs, thereby constituting a multi-objective problem (MOP). Central to our investigation is the identification of optimal values for key growth parameters, encompassing carbon dioxide (CO2) concentration, light intensity (LI), duration of light exposure (DL), specific Centella Asiatica Urban (CAU) cultivars or strains (SC), and the balanced distribution of different light spectra within the growth environment (RL).To achieve our objectives, we have develop a new hybrid approach combining statistical methods and metaheuristics is proposed to solve the MOP. We integrate a D-optimal design, multiple regression, and an innovative artificial multiple intelligence system (AMIS). This study, makes a significant contribution to the theory and practice by introducing a comprehensive and species-specific cultivation model, using as an example CAU.Through rigorous comparative analysis, the advantage of the AMIS algorithm over genetic algorithms (GA) and differential evolution algorithms (DE) is evident, resulting in substantial yield improvements. Our findings demonstrate an enhancement of 14.27