Process planning is a time-consuming and tedious operation that is necessary to manufacture products. In addition, process plans must be optimized to drive machines most productively. To accomplish this goal, we search for plans that satisfy production feasibility requirements and time constraints. Unfortunately, this search operates in an enormous search-space that is usually impossible to navigate. Therefore, we use features and their corresponding heuristics to guide the search. To make the search more tractable, we convert some features into corresponding constraints as a prior knowledge to prune the space. The applications of these features are presented in this paper. We also discuss the feature interaction problems in the process planning.

In this paper, we focus on the production of bent sheet metal parts and the machine setup. We have designed over forty test parts with our ``Parallel Design'' system, and the features are automatically generated as the design progresses. After the design is complete, an automatic process planning system uses the features and generates new ones to aid the production of a near-minimum manufacturing cost plan. Finally, these plans are used to produce parts on an automatic bending system.

C.-H. Wang and D. A. Bourne, ``Using Features and Their Constraints to Aid Process Planning of Sheet Metal Parts,'' in Proc. IEEE. Int. Conf. on Robotics and Automation, Vol. 1, pp.1020-1026, 1995.