<p>Drying is a critical process in various industries such as agriculture, food processing, and textiles, where maintaining precise temperature and humidity is essential for product quality and shelf life. Traditional methods often struggle with inefficiencies due to the interdependence of temperature and humidity. This study proposes a novel hybrid control approach that integrates the Reptile Search Algorithm and Random Forest Algorithm to optimize the variables of a Fractional-Order Proportional-Integral-Derivative controller. Additionally, an Adaptive Neuro-Fuzzy Inference System is used to model the relationship between temperature and humidity. Implemented in MATLAB/Simulink, the approach was evaluated across four testing scenarios: constant, step, saw tooth, and random responses. The findings show that the proposed strategy works significantly outperforms traditional techniques by reducing errors and maintaining more stable temperature and humidity control. For food dryers, it achieved humidity stability from 0 to 70%, with grain dryers reaching peak humidity of around 90% and wood dryers maintaining lower humidity levels. Key performance improvements include faster rise times, reduced overshoot, and lower steady-state errors. These findings illustrate the efficiency of the proposed approach in enhancing drying efficiency and system stability.</p>