Attributes charts are commonly used in monitoring quality characteristics of the proportion type and these charts assume that the monitored characteristics are binomially distributed. Classical control charts need to certain and precise data. However, in practice, quality experts express their opinion in imprecise form, which in turn, add more uncertainty and ambiguity. It is essential to properly represent and interpret uncertain information to evaluate product items. In this paper, the evidential reasoning (ER) based approach has been developed for supporting this uncertainty. So the belief multinomial p-control chart is introduced for monitoring the production process in the uncertainty condition, using evidence theory and the belief structure. A numerical example showing production process evaluation is examined by using the ER approach. The results show the proposed approach, is effective not only for reducing production defective but also for increasing the certainty in interpreting of quality variables (data).