This study introduces an innovative adaptation of response surface methodology (RSM) by implementing piecewise modeling

to address the limitations inherent to traditional second-order polynomial models. Traditional RSM often struggles

with complex, nonlinear system behaviors, particularly when variable interactions exhibit abrupt changes or asymmetrical

relationships. By segmenting the response surface into distinct regions, each modeled separately, the piecewise approach

enhances the methodology's adaptability and accuracy in predicting complex system dynamics. The eff ectiveness of the

proposed piecewise RSM is demonstrated through case studies, including the optimization of tetracycline removal from water

using a combined adsorption-coagulation process. This approach not only improves prediction accuracy but also integrates

economic considerations into process optimization, which is crucial for industrial applications where cost-eff ectiveness

is as important as operational effi ciency. The results indicate that piecewise RSM can provide more accurate modeling of

environmental and chemical engineering processes, providing a robust tool for improving experimental designs and process

effi ciencies while maintaining its simplicity.