Design for manufacturability methodology and data representation framework for machined components
MetadataShow full item record
The traditional product development process has been sequential in nature, with the product going through design, process planning, manufacturing and assembly. This sequential decision making results in increased costs and higher product development times. With the trend towards better product quality, product customization, shorter product life cycle, and international competition, manufacturers are faced with the challenge of improving product quality while reducing product development time, manufacturing lead-time, and product cost. To cope with these challenges, the product development process has to be made more efficient by integrating manufacturing and assembly considerations in the design phase itself, through the use of techniques such as Design For Manufacturability (DFM) and Design For Assembly (DFA).
DFM techniques have to be automated to take advantage of the vast advances in CAD and CAM systems. However, the automation of DFM has been constrained, especially for machined components, by the lack of methodologies which are dependent on the process of manufacture, and the incomplete part data representation in CAD systems. This research created a DFM methodology for machined components, along with an appropriate data representation scheme. Also, a software prototype was developed to demonstrate and validate both the methodology and the data structure.
The DFM methodology consists of three modules: DFM feasibility, process plan generation, and DFM analysis. The DFM feasibility module performs an initial feasibility check on the material, dimensions, tolerances, and configuration of the part. It also generates the spatial relationships between features. The process plan generation module uses a sequence identifier algorithm to generate the manufacturing sequence. The DFM analysis module evaluates tolerances relative to their stacking effects and manufacturability. It then analyzes the part configuration for possible design and process plan improvements.
A software prototype was developed using C++. It addresses the dimension checking, tolerance checking, configuration checking and spatial relationships generation in the DFM feasibility module. In the process plan generation module, the sequence of surfaces/features to be generated has been automated. This sequence is one of the major inputs to a computer-aided process planning module.
Other methodologies for non-machined components can be easily integrated into the DFM framework for complete automation of DFM analysis.
- Doctoral Dissertations