Manufacturing error analysis and Monte Carlo simulation for proactive tolerance control in process planning

In discrete part manufacturing, the quality of finished components is mainly controlled by specifying design tolerance requirements. Manufacturing engineers are then faced with the problem of selecting appropriate machining processes and production equipment to ensure that design specifications are satisfied. This is the task of process planning. Developing a process plan suitable for release to production is a complicated and time-consuming process that may take weeks or even months, where physical trial-and-error runs and many iterations of modification are needed. How many test runs and iterations are needed depends on the complexity of the parts and the experience of the planner.;In order to overcome the inefficiency of such a physical trial-and-error process in creating a process plan, this research aims to construct a system to simulate the procedures prescribed in a process plan through the prediction of component dimensional and geometric accuracy after each operation. The proposed system has two functions: (A) Through Monte Carlo simulation, performing manufacturing error synthesis, and predicting the overall manufacturing error effects in terms of dimensional and geometric tolerance. (B) Creating a systematic framework for process plan verification and enhancement.;The proposed error synthesis mechanism differs from traditional manufacturing error analysis in that it is based on actual deviation of a part's spatial locality caused by manufacturing errors and can potentially takes into account every kind of error source that contribute to surface inaccuracy, in contrast to the usual manufacturing error analysis which does not aggregate all possible error factors. A set of sample points are used to represent the dimension and geometry of an actual workpiece. During machining, various error sources will drive these sample points from their ideal position to non-ideal positions and as a result, an inaccurate part is developed. These erratic sample points will be studied and simulated using proposed error synthesis algorithm and Monte Carlo simulation. The inaccuracy of a workpiece is traced by an error-tracking system while it is propagated based on the process plan through subsequent machining operations, until the finished component is obtained. Virtual inspection is proposed to evaluate how much the sample points deviate from their ideal position to predict tolerances. The predicted tolerances can be compared with the design specifications to see whether the desired product tolerances are achieved by the given process plan.;Process planning is required in any manufacturing enterprise, regardless of plant size, part complexity and batch size. Any improvement in developing a successful process plan significantly aids the manufacturing industry. This research contributes to create opportunities to reduce the high cost of trial-and-error method by predicting tolerances, accomplish proactive tolerance control, and provide a rigorous base to tolerance analysis.