Abstract: At present, the digitalization and intelligence level of dripping pills production process is low, and there is a lack of process monitoring methods, which makes it difficult to effectively control the quality of dripping pills. Therefore, this paper proposed an online monitoring method for the dripping process of dripping pills based on laser detection technology and multivariate data analysis (MVDA) technology. Firstly, the width data of the falling droplets during the dripping process of the dripping pills were collected by the laser detector at a high frequency. Secondly, based on the width data, the nodes were selected for each droplet and the features were extracted. Then, the principal component analysis (PCA) model was established based on the feature dataset under normal process conditions, and Hotelling's T² or DModX statistic was selected to determine whether the droplets in the dripping process were abnormal, and the abnormalities were classified and diagnosed by the principal component score map combined with K-nearest neighbor (KNN) algorithm. In this study, the feasibility of this method was investigated by taking the dripping process of Ginkgo biloba leaf dripping pills as an example. The results showed that the obtained model has good detection and diagnosis ability for abnormal valve opening, abnormal liquid temperature, and abnormal liquid volume. This method can provide some reference for the industrial production of dripping pills.