Generative Approach

In the generative approach process plans are generated by means of decision logics, formulae, technology algorithms, and geometry based data to undertake a large number of processing decisions for converting a part from raw material to a finished state. The rules of manufacturing and the equipment capabilities are stored in a computer system. When using the system, a specific process plan for a specific part can be generated without any involvement of a process planner. For generative systems, input can come either as a text input where the user answers a number of questions in an English or English-like dialogue (defined as interactive input), or since graphic input where the part data is obtained from a CAD module (defined as interface input).

So far, the former procedure is more common in existing CAPP systems, while the later is still a fairly undeveloped area due to its complexity. Neverthless, interface input is necessary to enable an integrated manufacturing system. It has attracted much effort to interface CAPP with CAD. The terms, feature recognition, feature extraction, feature refinement, and geometry reasoning have been used to denote the study which will be discussed in detail later in this unit. Tulkoff states that 'Generative process planning systems today are still somewhat elusive on the whole and can be considered as being in their early stages of development and use (Tulkoff 1987).

The generative approach is complex and a few years back generative CAPP system is was too complex to ever be computerized. However, with the rapid development of AI techniques, the success of applying AI techniques in other areas has greatly encouraged the utilization of the AI techniques in the process planning. This effort has given initial results which indicate that the generative systems are desirable and promising.