Initial Development Approach In the following eight subsections, we present a brief discussion of methodology for expert system development, selection of problem and tools, knowledge engineering and prototype implementation, operational feasibility, and the actual development of a working prototype of a process planning expert system. Methodology for Expert System Development Expert systems require a software development methodology that differs in some respects from those methodologies used for conventional systems. Most knowledge-based development methodologies used by organizations experienced in building expert systems are similar in that they concentrate on the early (feasibility) stages of a project. Very little has been published on the later stages, which are concerned with expert system delivery, integration, and maintenance. During the development of OPGEN, we incorporated the lessons learned in these early stages and revised our original approach to provide for integration and maintenance. Most expert system development methodologies are a variation on the following theme, which paraphrases Haycs-Roth (1985): (1) expert system technology is determined to be relevant to a product; (2) management provides an opportunity for action; (3) a preliminary business application is assembled; (4) a knowledge engineering consultant verifies the opportunity; (5) a knowledge engineering project team is formed and assesses the knowledge; (6) the knowledge engineering project manager plans the project; (7) the user organization Figure 2 OPGEN bzput Circuit Layout Diagram.

Large engineering projects, such as the engineering development of computers, involve a large number of activities and require cooperation across a number of departments. Due to technological and market uncertainties, these projects involve the management of a large number of changes. The Callisto project was born out of realization that the classical approaches to project management do not provide sufficient functionally to manage large engineering projects. Callisto was initiated as a research effort to explore project scheduling, control and configuration problems during the engineering prototype development of large computer systems and to devise intelligent project management tools that facilitate the documentation of project management expertise and its reuse from one project to another. In the first phase of the project, rule-based prototypes were used to build quick prototypes of project management expertise and the project management knowledge required to support expert project managers. In the second phase, the understanding of point solutions was used to capture the underlying models of project management in distributed project negotiations and comparative analysis. This article provides an overview of the problems, experiments, and the resulting models of project knowledge and constraint-directed negotiation.

The Parametric Interpretation Expert System (PIES) is a knowledge system for interpreting the parametric test data collected at the end of complex semiconductor fabrication processes. The system transforms hundreds of measurements into a concise statement of all the overall health of the process and the nature and probable cause of any anomalies. A key feature of PIES is the structure of the knowledge base, which reflects the way fabrication engineers reason causally about semiconductor failures. This structure permits fabrication engineers to do their own knowledge engineering, to build the knowledge base, and then to maintain it to reflect process modifications and operating experience.

The Parametric Interpretation Expert System (PIES) is a knowledge system for interpreting the parametric test data collected at the end of complex semiconductor fabrication processes. The system transforms hundreds of measurements into a concise statement of all the overall health of the process and the nature and probable cause of any anomalies. A key feature of PIES is the structure of the knowledge base, which reflects the way fabrication engineers reason causally about semiconductor failures. This structure permits fabrication engineers to do their own knowledge engineering, to build the knowledge base, and then to maintain it to reflect process modifications and operating experience. The approach appears applicable to other process control and diagnosis tasks.

EDITORIALS This Fall issue marks the first time we have devoted the AI Figure 1 summarizes the results of a survey I ran in 1985. The idea originated a couple of depicts the number of AI based systems in the various stages years ago, and I'm pleased to see the actual implementation. of research, development, field service and production use. Mark Fox, Special Editor for this issue, is to be congratulated It is my guess that the survey represents about 15% of the for a fine job of selecting some of the best authorities in systems currently under development. The incursion of AI the field and working with them to produce an excellent survey into the manufacturing world has reached the point that discussions of the current state of the art in AI for manufacturing. The quality of all the articles was so high that we didn't want to exclude any of them.

Our development efforts involve small multidisciplinary KLA Instruments Corporation is looking for bright, dedicated teams with backgrounds in CS, EE, Math, Mechanical professionals to meethe challenge of developing Engineering and Physics, oriented to building leading edge knowledge-based systems for machine successful commercial products. You will have the control. We're using existing Al technology to actually opportunity to work on entire life cycle development get new KLA products out the door. If you have experience from idea to first shippable products.

The development of an online turbine generator diagnostic system is described from conception to initial field verification. The system is composed of a data center located in the power plant that collects data from online measurement devices and communicates these data to a centralized diagnostic facility in Orlando, Florida, where the actual diagnosis is done. The resulting diagnosis and recommended actions are transmitted to the power plant where they are displayed to the operator by the data center. The market-place need, initial approaches to the product, system field verification are described. The artificial intelligence (AI) diagnostic program has been diagnosing seven large utility generators since July 1984 and has correctly diagnosed a significant number of generator and instrumentation problems. Issues such as a centralized approach, rule base quality control, and the range of resources needed for a successful product are discussed.

This article provides a discussion of factory design and an artificial intelligence (AI) approach to this problem. Major issues covered include knowledge acquisition and representation, design methodology, system architecture, and communication. The facilities design expert systems (FADES developed by the author is presented and described to illustrate issues in factory design.

This report describes recent and continuing research in artificial intelligence and related fields being conducted at the University of Pennsylvania. Although AI research takes place primarily in the Department of Computer and Information Science ( in School of Engineering and Applied Science), many aspects of this research are preformed in collaboration with other engineering departments as well as other schools at the University, such as the College of Arts and Sciences, the School of Medicine, and Wharton School.

This is the second part of a two-part article describing AI work at the NASA Johnson Space Center (JSC). Research and applications work in AI is being conducted by several groups at JSC. These are primarily independent groups that interact with each other on an informal basis. In the Research and Engineering Directorate, these groups include (1) the Artificial Intelligence and Information Sciences Office, (2) the Simulation and Avionics Integration Division, (3) the Avionics Systems Division, and (4) the Tracking and Communications Division. In the Space Operations Directorate, these groups include (1) the Mission Planning and Analysis Division (MPAD) - Technology Development and Applications Branch, (2) the Spacecraft Software Division, and (3) the Systems Division - Systems Support Section. This second part of the article describes the AI work in the Space Operations Directorate. The first part of the article, published in the last week of AI Magazine, (7:1, Summer 1986) described the AI work in the Research and Engineering Directorate.