A comprehensive rapid inspection experiment for students majoring in quality management engineering is designed. The experiment uses a portable spectrometer to quickly collect sample quality characteristic data, and then uses a quality statistical tool and an intelligent learning algorithm to achieve rapid data analysis and pattern revelation. By irradiating dairy products with a Raman spectrometer, scattering characteristic signals of the samples are obtained, and the data collection for each experimental sample only takes 60 seconds. Feature peaks are selected and quality statistical control analysis tools are uesd to draw a sample quality control chart, which can describe the fluctuation of sample quality. Furthermore, by combining an extreme learning machine algorithm with data preprocessing methods, the rapid identification of sample brands can be optimized, with a recognition rate of 97.3% and an algorithm running time of only 1 s.