Product quality is a key factor for manufacturing companies to evaluate their production capability and increase their market competitiveness. Today's Manufacturing processes have become more complicated and usually equipped with smart sensors that collect a massive amount of data along the manufacturing chain. This chain consists of a multistage of manufacturing processes to produce complex products to satisfy customer requirements. In multistage manufacturing systems, many factors may have interactive and cumulative effects on the final product quality. The purpose of this research is to introduce an intelligent real-time quality monitoring framework capable of predicting and identifying the quality deviations for multistage manufacturing systems as early as possible to reduce wastes of time and resources. We used different unsupervised and supervised machine learning techniques such as principal component analysis, support vector machine, neural network and random forest to consider the accumulative effect of different workstations and to construct the quality monitoring model. We used a complex semiconductor manufacturing dataset to evaluate the performance of the proposed framework. The results show the capability of the proposed framework to improve the performance of the quality monitoring process in the multistage manufacturing systems and to reduce both type Ӏ and type ӀӀ errors.