Progressive dies for producing sheet metal parts in mass production ave been widely applied in various industries, such as aerospace, electronics, machine tools, automobiles, and refrigeration. These dies can perform piercing, notching, cut-off, blanking, lancing, bending, shaving, drawing, embossing, trimming,coining, and other miscellaneous forming operations at a single setup. Hence a progressive die is generally very complex. Stamping process planning and die structure design are difficult and demanding tasks.

Stamping process planning starts with an unfolding of a model of stamped metal to produce a flat pattern, followed by nesting the pattern to produce a blank layout. Next, stamping operations are planned and operations are assigned to die stations. The resulting plan is typically represented as a strip layout, which guides the subsequent progressive die structure design. The productivity, accuracy, cost and quality of a progressive die mainly depends on the strip layout, and hence a stamping process. However, stamping process planning still remains more of an art rather than a science. Historically, this activity is mainly carried out manually, based on designers’ trial-and-error experience, skill and knowledge.

To address the above issue, it is necessary to provide a cooperative problem solving strategy that can foster communication between diverse KS’s, and accommodate different knowledge representation schemes within an integrated framework. For this purpose, a hybrid intelligent system consisting of a blackboard control module and a few independently executing KS’s is proposed. This hybrid system provides a cooperative decision making environment and facilitates a hybrid knowledge representation scheme, including procedures, production rules, object-oriented, graph-based and case-based representations. The proposed approach speeds up the progressive die design process by automating the strip layout design.

Recent advances in the field of artificial intelligence (AI) have given rise to the possibility of constructing AI-based system that incorporate built-in intelligence and apply diverse knowledge to solving progressive die design problems, including strip layout design automation. The diverse knowledge sources(KS’s) related to stamping process planning include unfolding knowledge to produce a flat pattern, nesting knowledge to produce a blank layout, various types of planning knowledge for different stamping operations like piloting, piercing, notching, cut-off, blanking, bending, etc., and staging knowledge to sequence the stamping operations. However, the existing work is based on the conventional architecture of knowledge-based expert system, which are incapable of managing heterogeneous KS’s effectively. In addition, these work doesn’t provide a representation scheme for experts to model their valuable, but difficult-to-articulate, knowledge in terms with which they are familiar.

Owing to the modeling flexibility of blackboard architecture, our earlier work on case-based reasoning (CBR) for stamping process planning and die design is also integrated to facilitate the search and reuse of past design experience to solve new problems. The limitation of a single CBR approach has been overcome in the proposed hybrid intelligent system, which can employ other reasoning approach such as rule-based reasoning for problem solving where relevant data have not yet accumulated in the case base.