Traditional attribut control charts judge the product quality by the binary classification. This is not appropriate in many situations where product quality does not change abruptly from satisfactory to worthless, and there might be a number of intermediate levels. Not fully utilize the intermediate information results the in poorer performance than -chart. Previous research proposes that the intermediate levels may be expressed in the form of linguistic terms and then to use the fuzzy set theory to construct the fuzzy attribute control charts to monitor the data in linguistic form. In this research, we propose a neural network-based quality control procedure to improve the performance of the fuzzy control charts. The primary issue focuses on the detection of changes in the process mean shift. The imput of the neural network is linguistic variable. We try to us the investigation of design stratgies to improve the detecting. The neural network performance is evaluated by on average run length. An extensive simulation study indicates that the proposed neural network approaches is better than fuzzy attribute control charts.