This thesis examines geometric process planning in four-axis rough machining. A review of existing literature provides a foundation for process planning in machining; efficiency (tool path length) is identified as a primary concern. Emergent structures (thin webs and strings) are proposed as a new metric of process robustness. Previous research efforts are contrasted to establish motivation for improvements in these areas in four-axis rough machining.

The original research is presented as a journal article. This research develops a new methodology for quickly estimating the remaining stock during a plurality of 2 y D cuts defined by their depth and orientation relative to a rotary fourth axis. Unlike existing tool path simulators, this method can be performed independently of (and thus prior to) tool path generation. The algorithms presented use polyhedral mesh surface input to create and analyze polygonal slices, which are again reconstructed into polyhedral surfaces. At the slice level, nearly all operations are Boolean in nature, allowing simple implementation. A novel heuristic for polyhedral reconstruction for this application is presented. Results are shown for sample components, showing a significant reduction in overall rough machining tool path length.

The discussion of future work provides a brief discussion of how this new methodology can be applied to detecting thin webs and strings prior to tool path planning or machining.

The methodology presented in this work provides a novel method of calculating remaining stock such that it can be performed during process planning, prior to committing to tool path generation.