Feature-based modelling and neural networks-based CAPP for integrated manufacturing

Abstract

The integration of various technical activities in a manufacturing organization using computer-based systems is very effective in improving its competitiveness. ComputerAided Process Planning (CAPP) is an essential function in Computer-Integrated Manufacturing (CIM). During process planning, the design information is interpreted and appropriate manufacturing operations specified, and this activity requires a significant amount of expertise. In this paper, a methodology of integrating the design and planning aspects of manufacturing is proposed which utilizes the concepts of Feature-Based Modelling (FBM) and Artificial Neural Networks (ANNs). Important process planning activities are identified for the application of neural networks in realizing intelligent CAPP systems. An intelligent CAPP methodology is developed to suggest the appropriate manufacturing operations and their subdivisions. It is then integrated with an FBM shell, which is used to represent the design aspects of the component from a manufacturing point of view. An example is presented. The use of ANNs approach has enabled the development of a very flexible CAPP methodology that can easily be trained to handle new types of components.