Extraction of milling know-how from NC programs through reverse engineering

Abstract

To automate machining process planning, the acquisition and representation of machining knowledge or know-how in a reusable way is needed. The machining know-how is implied in NC programs made by experienced workers. In this paper, the methodology and system for extracting machining know-how in milling operations have been developed. With the system, machining features, operations and their associated cutting conditions (depth of cut, feed rate and spindle speed, etc.) and machining method can be extracted by analysing NC programs in conjunction with the tools used and workpiece blank model. The milling know-how is represented as a collection of these extracted data that can be used in future machining operations. To verify the system, actual NC programs for milling have been analysed and the milling know-how has been extracted successfully. Introduction Existing manufacturing de® ciencyAs the manufacturing paradigm is shifting from skill based to knowledge based, and operation is shifting from a local to a global scale, manufacturing expertise becomes more and more critical in sustaining the competitive manufacturing operation. As the experienced workforce is decreasing nowadays while new operators still do not have enough expertise to fully utilize the functions built in today's highperformance CNC machine tools, the manufacturing know-how of these skilled workers needs to be retained as intelligence in future manufacturing (Dong and Vijayan 1997).On the other hand, as machine tools evolve to achieve even higher performance (higher productivity and accuracy), they require highly skilled operators for full utilization of their performance. Moreover, some high-speed CNC machine tools cannot securely and e ciently be operated even by the skilled operators because they require careful and innovative operation planning schemes. Such an operation planning method requires new and sophisticated machining principles that have not yet been experienced by conventionally experienced machine tool operators and production engineers (Chan and Yamazaki 1997). Without a reliable manufacturing knowledge or know-how database, state-of-the-art machine tools cannot be fully utilized.In order to achieve highly e cient manufacturing, CNC machining based on CAD

Reverse engineering of machining operation planning

“…To extract the machining know-how data from NC programs, a system called Autonomous Machining Process Analyzer (AMPA) has been developed (Chan and Yamazaki 1997, Yan et al 1999, 2000a. This system is used to extract the machining knowledge or know-how of the skilled operators and machinists, which is implied in successful NC programs, and automatically export it into the know-how database.…”

Development of Basic System to Construct Database of Machining Know-how

“…In addition, operators without the required skills are unable to easily output sophisticated NC programs for desired product shapes on existing CAM systems. Yan [4] proposed a similar system for extraction of machining know-how from NC programs. However, the system can not deal with cutting condition changes in an NC program done by experienced engineers at the manufacturing site.…”