Pamela K. Fink

Expert Systems and Diagnostic Expertise in the Mechanical and Electrical Domains

Current expert system technology tends to rely on the use of shallow empirically based experiential knowledge. With only this type of knowledge available, expert systems have been capable of reaching a high level of agreement with human experts in a limited area of expertise. However, due to the nature of their knowledge, such systems fall short of human expertise in many ways. The human diagnostic process is examined as it relates to the malfunction of mechanical and electrical devices. An expert system design is presented, called the integrated diagnostic model (IDM), that attempts to address some of the issues involved in bridging the gap between human and computer expertise. The IDM contains two different types of knowledge, one based on experience and one based on how the device to be diagnosed functions. These two types of knowledge are used together during a diagnostic session to determine what is wrong with the device. To demonstrate how the IDM works, an interaction with a prototype system that was built using the IDM is described; then research on extensions to the IDM is discussed.

A General Expert System Design for Diagnostic Problem Solving

Existing expert systems have a high percentage agreement with experts in a particular field in many situations. However, in many ways their overall behavior is not like that of a human expert. These areas include the inability to give flexible, functional explanations of their reasoning processes, and the failure to degrade gracefully when dealing with problems at the periphery of their knowledge. These two important shortcomings can be improved when the right knowledge is available to the system. This paper presents an expert system design, called the integrated diagnostic model (IDM), that integrates two sources of knowledge, a shallow, reasoning-oriented, experiential knowledge base and a deep, functionally oriented, physical knowledge base. To demonstrate the IDMs usefulness in the problem area of diagnosis and repair, an implementation in the mechanical domain is described.

Image Processing and Artificial Intelligence for Detection and Interpretation of Ultrasonic Test Signals

Detection of flaws is an important industrial concern. For example, aircraft and nuclear-power reactor owners and regulatory authorities need effective means of detecting flaws that could pose a threat to public safety. Operators of costly equipment require information on service-induced flaws to be able to make run-or-retire decisions. As the cost of parts and concerns for public safety increase, the importance of flaw detection and size estimation has likewise escalated.

The disease progression explorer: risk assessment support in chronic, multifactorial diseases

Managing and treating chronic, multifactorial diseases, such as atherosclerosis, osteoporosis, asthma, and periodontal disease, requires understanding the role of each risk factor in the disease and the interrelationships between risk factors. The discovery of a new risk factor often generates a new way of looking at the disease process and how to anticipate its expected progression. The Disease Progression Explorer concept was designed to support the transfer of information about a new risk factor for a disease from the research lab to clinical practice. Through the use of the Disease Progression Explorer, this transfer is supported by a graphical system that shows directly how risk factors, alone and in combination, influence the expected course of the disease and shows how disease management strategies and therapies can affect the patient’s anticipated outcome. This paper describes a Disease Progression Explorer for periodontal disease called the Perio-DPE.

Thinking, decision making and problem solving

The stat-of-the-art in the development of intelligent systems has come a long way over the 30 year history of artificial intelligence. Techniques have evolved from the development of more general purpose problem solvers to the recognition that human problem solving which is highly dependent on lots of specific knowledge. Thus, approximating human problem solving capabilities requires the acquisition and input of a large amount of knowledge specific to the particular problem solving task. To develop such systems requires techniques, either manual or automatic, to acquire, represent/store, and utilize the knowledge needed to perform a selected problem solving task in a computer software program. These challenges are being addressed in a variety of ways from both an applications and a research perspective. Several example approaches developed and/or utilized at Southwest Research Institute have beers summarized.<<ETX>>

An intelligent facility planning adviser

Abstract The problem of facility planning and allocation of floor space to specific functional work groups in an industrial environment requires many different pieces of information to be considered. In order to reduce the complexity of the problem-solving task and to achieve a reasonable decision in a timely manner, facility planning is often based on extensive general and approximate knowledge of the data required and a lot of experience with the given facilities, workgroups, products, and processes. In order to make more accessible the data needed to make decisions concerning use of facilities, as well as to support execution of a formal and effective methodology for handling the wide variety of activities concerned with the problem of facility planning, an artificial intelligence (AI)-based system, called the Intelligent Facility Planning Advisor, has been under development for the Air Force Air Logistics Center at Hill Air Force Base in Ogden, Utah. This paper describes the first two components developed for this system, called the Intelligent Floor Allocation Advisor and the Intelligent Department Level Layout Planning Advisor.