MLR and PLS models with or without OSC were studied through establishing quantitative calibration models for the peanut oil content in blending edible oils, and for the dimethylsulfoxide concentration in water solution. The cross validation results and the predication results of MLR models, were compared to evaluate the effectiveness of OSC for improving the performance of MLR model. The results show that the SEC or SEP of MLR models using OSC gets smaller. Selecting appropriate wavelengths combination by CARS method, prediction capacity of MLR model using OSC is better than PLS1 model using raw spectrum.