To improve the precision and robustness of the model for the chlorpyrifos active ingredient in pesticide EC by near infrared spectroscopy,synergy interval PLS(siPLS) combined with genetic algorithm(GA) was implemented to optimize the feature variables,and cross-validation method was used to select the optimal PLS factors and the variables.The results showed that the optimal model was achieved with R2p of 0.972,root mean square error of prediction(RMSEP) of 0.353% in the prediction set when 81 variables and 11 PLS factors were included.Experimental results showed that siPLS combined with GA could eliminate a large margin of the redundant information and irrelevant information in pesticide EC spectroscopy,and reduce the complexity of the developed model.The precision and robustness of the model were also improved.