In the environment of industrial big data, the observation has high-dimensional, low value density and outlier. It is difficult to dig the important variation information from mass data, which will lead to a high false alarm rate and affect the product quality. To overcome this problem, a novelty robust monitoring model in the high-dimensional process was proposed based on minimum covariance determinant estimation and variable selection algorithm. First, the mean vector and covariance matrix can be obtained by MCD method; Second, the objective function of variable selection algorithm can be modeled by adding a penalty term, which can be applied to recognize the potential abnormal variables; Third, the final monitoring statistic can be established, and the control limits can be calculated by Monte Carlo method; In the last, the simulation and real example were applied to verify the performance of proposed method. The result shown that the proposed robust variable selection control chart outperforms than Hotelling T2 and conventional variable selection control chart, because the proposed chart can detect the outliers from the massive observations, which makes the monitoring model more robust.