A Meta Knowledge Base for Machining Process Selection
Abstract The work reported here is part of a larger project aimed at developing a system for concurrent design and manufacturing. One of the manufacturing applications is manufacturability evaluation performed on design models based on features. The model is not restricted to a limited set of predefined features, but can contain user-defined features that the application program, such as the Manufacturability Evaluator, has no prior knowledge of. A methodolgy was developed whereby the manufacturability evaluator could understand design features from a manufacturing viewpoint and automonously generate feasible machining process sequences for undocumented features. This requires that the system decompose design features into some generic fundamental terms and compare this representation to the capabilities of common machining operations. Therefore, the system needs deeper knowledge about machining processes than that afforded by production rules. In our method, undocumented features are partially evaluated to derive the faces intersecting at concave edges. An algorithm is developed to organize these faces into manufacturing features. The meta knowledge base captures the fundamental characteristics of a machining process by its elementary producible volume and the limitations of tool motions. This representation enables the manufactuability evaluator to create alternative machining sequences for undocumented features.