The Basis and Needs for Intelligent Materials Processing
- PDF / 4,177,912 Bytes
- 4 Pages / 604.8 x 806.4 pts Page_size
- 48 Downloads / 192 Views
material when the total processing is completed. In fact, only in the light of a definition of the properties and the processes is it possible to define the parameters that can be monitored for on-line control. The three main components of an intelligent processing system are shown in Figure 1. The word "intelligent" is used to connote that the system is constantly learning, and that each occurrence is rationalized or evaluated in comparison to what is stored in the knowledge bank; this concept is shown in Figure 2. The benefits of incorporating intelligent processing methods in the manufacturing cycle can be appreciated when one compares the practice of quality assurance with quality control. The former ensures that throughout each step of the process the measurements sensed are in-line with the ideal, and if deviations are noted, changes are made. Quality control ensures that the produced component meets the established specifications. The component is scrapped if it does not meet the specifications; a quality product is shipped at the expense of productivity. The components of intelligent materials processing are briefly described below.
Process Understanding and Models Process modeling is the foundation of process design and implementation. The ideal process model begins with specification of starting materials and ends with structure and composition to Editor's Note: This article is an abridged version of permit the prediction of final properties. the one originally published in Materials Futures: The full implementation of process Strategies and Opportunities, Proceedings of modeling in the design and control of the U.S.-Sweden Joint Symposium held October materials processing is at least a decade 18-19, 1988, Philadelphia, Pennsylvania (Mateaway. Current approaches to process rials Research Society, Conference Proceedings, design and implementation are not 1988) p. 107-115.
24
always useful because of a lack of fundamental knowledge and commercially available equipment. A new approach is needed to ensure that on-line control does not degenerate into merely a proliferation of sensor installations, resulting in complexity and fault intolerance rather than simplicity and quality. A manufacturing process is based on a prescription of activities that transforms a given set of raw materials into the final product. This prescription consists of a mixture of empirical and fundamental operations. In fact, the most critical frontier in the science of materials today is this boundary between the ability to make things and the fundamental understanding of the processes. To compete successfully in the cost and quality of modern manufacturing, the practices of the past must be reconsidered; the total sequence of processes must be integrated and standardized; and on-line control must be practiced at critical steps. A decision regarding implementation of on-line control versus post-process testing can be made based on the complexity and real cost of the individual process step. Optimization for each materials system and
Data Loading...
