A quantitative structure activity relationship (QSAR) model for a series of N-(1-benzyl-3,5-dimethyl-1H-pyrazole-4-yl) benzamide derivatives having autophagy inhibitory activities as potent anticancer agents was developed by the multiple linear regressions (MLR) method. In this study, previous compounds were used in the model development were divided into a set of fifteen compounds as training set and set of four compounds as test set. A model with high prediction ability and high correlation coefficients was obtained. This model showed r = 0.968, r2 = 0.937 and Q2 = 0.880, the QSAR model was also employed to predict the experimental compounds in an external test set, and to predict the activity of a new designed set of 3,5-dimethyl-4-substituted-pyrazole derivatives (1-15), result showed that compound 3 has the most promising inhibition activity (EC50 = 0.869 μM) against human pancreatic ductal adenocarcinoma cell MIA PaCa-2 compared to the reference chloroquine with (EC50 = 14 μM). Thus, the model showed good correlative and predictive ability. Docking studies was performed for designed compounds, docking analysis showed the best compound 1 with high docking affinity of -24.8616 kcal/mol.