Research Initiation: Neural Networks and Fuzzy Control Application for Quality Engineering in Short Run Manufacturing Environments

9410613常这项研究的重点是开发和使用模糊控制器，以解决小批量生产环境下的过程设计和控制问题。控制器将就给定生产实例中的机器操作参数的选择提供建议，以便生产出质量可接受的零件。在研究的初始阶段，将使用计算机仿真来测试控制器。随后，用于过程控制目的的其他现有模型产生的数据将用于测试和微调控制器的性能。在控制器研制成功后，将在计算机数控机床上设计和加工样件，模拟小批量生产环境，以进一步测试控制器。第二个目标是开发一种过程监控方法，该方法使用神经网络控制图(NNCC)来整合来自质量特征和与过程相关的信息的观察，以确定过程状态。研究成果对提高小批量生产条件下的工艺质量和工艺参数选择具有一定的工业应用价值。过程控制和过程设计的结合为过程数据收集提供了在线过程监测的机会，目的是通过减少过程操作差异来提高产品质量。

9410613 Chang This research is focused on the development and use of a fuzzy controller to address the problem of process design and control in a short run production environment. The controller will provide recommendations on the choice of machine operating parameter in a given production instance in order to produce a part of acceptable quality. Computer simulation will be used to test the controller in the initial phase of the research. Subsequently, data generated from other currently existing models for process control purposes will be used to test and fine-tune the performance of the controller. Upon the successful development of the controller, sample parts will be designed and processed on computer numerical control (CNC) machine to simulate a short run production environment to further test the controller. A second objective is to develop a process monitoring approach that uses neural network control chart (NNCC) to integrate observations from quality characteristic and process related information to determine process status. The outcome of the research has some potential for industrial application to improve process quality and process parameter selection for short-run production situation. The integration of process control and process design provides opportunity for on-line process monitoring for process data collection with the objective of improving product quality through reduction in process operational variation.