Abstarct: Alzheimer’s disease (AD) is one of the mosft severe neurodegenerative disorders and the leading cause of dementia in the elderly, characterized by progressive impairment in memory, cognitive functions, and neuronal structures. Due to the limited efficacy of current therapeutic agents and the multifactorial nature of the disease, the development of multi-target-directed ligands (MTDLs) has emerged as a promising strategy in modern drug design. In the present study, derivatives of ferulic acid—known for their antioxidant, anti-inflammatory, and neuroprotective properties—were investigated as potential multi-target agents. Quantitative structure–activity relationship (QSAR) models were developed to predict the inhibitory activity of these compounds against acetylcholinesterase (AChE). A dataset consisting of 49 ferulic acid derivatives was collected, and IC₅₀ values were converted to pIC₅₀ to be used as the dependent variable in the modeling process. Chemical structures were optimized using computational chemistry methods, and molecular descxriptors were extracted using specialized software tools. The modeling strategy employed a multistage STEPWISE-MLR-ANN (STP-MLR-ANN) frxamework. Initially, descxriptor selection was performed using the stepwise algorithm, followed by the construction of linear models via multiple linear regression (MLR). Subsequently, an artificial neural network (ANN) was developed to capture nonlinear relationships between molecular structure and biological activity. The resulting model exhibited high predictive accuracy and strong generalizability, as demonstrated by validation metrics including the coefficient of determination, RMSE, leave-one-out cross-validation (LOO), Y-randomization tests, and external test-set evaluation. The validated model was then applied to design new ferulic acid derivatives with enhanced predicted biological activity. These proposed molecules were further evaluated for their drug-likeness baxsed on Lipinski’s rule using the SwissADME platform. Finally, molecular docking studies were performed to analyze the binding interactions of the designed compounds with the active site of AChE. Docking results showed good agreement with QSAR predictions, confirming the potential of the designed compounds. Overall, this study demonstrates that integrating chemometric and computational chemistry approaches—including QSAR modeling, artificial neural networks, and molecular docking—provides a powerful frxamework for the predictive screening and rational design of ferulic acid derivatives. This strategy may facilitate the development of effective multi-target drug candidates for the treatment of Alzheimer’s disease.