A concurrent engineering approach for supporting weld product design at early stages of the design process

Traditional welding expert systems rely heavily on human design knowledge to obtain the necessary design information for welding process planning. Such systems can rarely be useful to support the concurrent engineering design process of weld products, especially during early design stages where most of the design information may still be unavailable. Also this problem cannot be solved by simply integrating traditional CAD systems with welding process planning systems, because these systems require a considerable amount of product data; operate in a sequential manner; and there is little "fusion" of welding expertise during the design configuration stage. It is thus necessary to develop a new generation of computer tools to support the design of weld products at the early stages of the design process. An aggregate product model (APM) is introduced in this paper, which is designed to be used for the design and process planning of weld products. The APM can be used for capturing essential product data in the early stages of design. Based on the APM, a concurrent engineering environment for weld product design and process planning is currently under development using an artificial intelligence programming environment and C language. The research has developed methods and algorithms for weld product configuration, welding position generation and selection, welding process selection and optimisation as well as welding process sequencing. These functions are discussed in this paper using simple design examples. The paper shows that the use of the APM in the design process can potentially improve the overall integrity of design and planning decisions by the automatic generation of compatible design alternatives and the rapid evaluation of these options to satisfy fabrication requirements.