An algorithmic strategy for automated generation of multicomponent software tools for virtual manufacturing

This paper describes an algorithmic strategy to facilitate the generation of multicomponent software tools for computer-aided manufacturing (CAM) and virtual manufacturing (VM). Components that are often used to build CAM and VM applications include a wide range of domain-specific knowledge sources and also more general utility components with often very heterogeneous characteristics. The identification of a suitable and compatible set of these components is the first and arguably most important step during the development process of any CAM or VM application. This paper presents an algorithmic strategy that automates this development step by solving a time-expanded network problem, referred to as the component set identification (CSI) problem. A definition of the CSI problem, a mathematical formulation, a solution procedure, and some computational results are presented. Finally, an application to predict hole quality in drilling is used to illustrate the functionality of the proposed algorithmic strategy.