In manufacturing industry the automated visual inspection system (AVIS) is a method to inspect, classify and detect defects of various products. In the past, the tasks of inspection are carrying out by humans, machines or both. In this paper, we account for an AVIS model to classify mechanical parts in production line. It comprises two parts: hardware and software. The model uses a web-camera attached to an adjustable stand to capture various group of metal part images. The main objective is to develop an intelligent inspection tool based on image processing and production rules. It computes both the area and circularity of mechanical shapes as the features and hence classifies them according to ten categories such as screws, nuts, and bolts at different sizes. The result shows that the accuracy is 91.5% for group and 98.25% for individual classification of mechanical parts subsequently.