Optimization of Complex Spray Drying Operations in Manufacturing Using Machine Learning: Evaluating Techniques for Energy Efficiency and Product Quality Enhancement

This paper investigates the application of machine learning techniques to optimize complex spray-drying operations in manufacturing environments. Using a mixed-methods approach that combines quantitative analysis with qualitative expert insights, the study demonstrates how algorithms can improve energy efficiency, product quality, and decision-making. A comparative analysis of Support Vector Machines, Bayesian methods, Decision Trees, and Ensemble techniques shows that ensemble methods, especially Random Forest yield superior predictive accuracy (R2 = 0.962), while decision trees enhance interpretability for operator support. The integration of algorithmic modeling with domain expertise produces robust optimization strategies by leveraging the strengths of both data-driven and human-informed approaches. The research contributes to the theoretical development of Statistical Learning Theory in the context of complex thermal systems and presents a framework for incorporating data science methodologies in Industry 4.0 manufacturing environments.