AI-Based Process Planning for Electronic Assembly

Diversity of electronic products, lack of human expertise, and inconsistency among manually generated process plans increase the necessity of developing Computer Aided Process Planning (CAPP) systems for electronic assembly. Artificial Intelligence (AI) offers a few techniques to develop CAPP systems which would behave in a 'knowledgeable' manner. Two rule-based CAPP systems, a backward-chaining system developed in M. 1 shell and a forward-chaining system developed in OPS 5, and an object-oriented system developed in Objective-C are discussed. The CAPP systems generate processes and select a set of machines to perform the generated processes for a given product. In the rule-based CAPP systems, product characteristics and bill of components are stored in working memory and capabilities of the available machines are stored as rules. In the object-oriented CAPP system, product as well as the available machines are represented as objects. A central decision module which does pattern matching and sends messages to appropriate machines (objects) is developed