Knowledge-based System Approach Research Articles

Human-computer interaction issues in developing a bearing design support system: a case study

This paper describes human-computer interaction issues within the development of an engineering design support system using a case study. The study takes a design support system for externally pressurized journal bearings as an exemplar. The issues presented focus on aspects of the system interface design and design interaction aids. The design principles of a system user interface are formulated from functional, aesthetic and structural perspectives. The principles are concerned with using metaphors, implementing direct manipulation, consistency design and aesthetic integrity, which are described in detail through the design of system interface elements such as icons, pull-down menus, screen layout and sound usage. The interaction aids for a designer using the system is discussed with particular reference to a knowledge-based system approach. The aids include the initial determination of values of design parameters, the monitoring of the computed parameter values and the final evaluation of the design outputs. With these aids, a designer can intuitively overview the design process and learn to control the process interactively. Human-computer interaction issues are important in establishing user satisfaction with system specifications such as the system interface, user-friendliness, performance speed and learning time.

Kinematic Structural Synthesis of Mechanisms Using Knowledge-Based Systems

This paper presents a knowledge-based systems approach to the automation of the conceptual synthesis of mechanisms. The system utilizes a procedure for synthesizing design alternatives based on the principle of separation of structure from function. This principle in turn resolves the problem of knowledge representation of design alternatives through the use of graph structures which are then evaluated using a set of heuristic rules. The expert system presented in this paper has been implemented and tested for the conceptual synthesis of variable-stroke engines and robot-hands kinematic structures. The system has shown the capability of producing numerous design alternatives for these two applications, and has provided consistent results based on the evaluation rules. It has also provided the designer with immediate feedback on the viability of any given design alternative and the reasons behind that decision.

Selection and sequencing of hazardous waste treatment processes: A knowledge-based systems approach

A commercially available software package was used to develop a prototype object-oriented knowledge-based modeling system to aid in the determination of hazardous waste treatment process sequence. For the prototype system, 15 waste contaminants were chosen for consideration. Knowledge describing 10 treatment processes was collected and incorporated into separate program modules so only those processes required were considered by the system. The knowledge-based system was established so that the user can specify the treatment objectives that the treatment sequence is intended to satisfy. The prototype system is intended to demonstrate a potential use of knowledge-based systems in both the design and the operational management of hazardous waste treatment facilities. From a design perspective, such a system could be used to determine effective treatment sequences for expected waste types, thus helping to determine which processes should be included in a proposed facility. From an operational management perspective, such a system could be used to determine the treatment sequence required for a waste to meet treatment objective contaminant levels.

Knowledge-based understanding of signals: An introduction

The paper provides an introductory review to the knowledge-based approach to signal understanding. Basic issues in knowledge-based system (KBS) design are presented following current methodologies, and peculiarities of interpretive problems are discussed in this context. Formal descriptions are given in terms of objects and tasks, the former denoting abstract data and concept representations, the latter generic knowledge-processing steps (numerical, inferential). Understanding patterns is seen as a sequence of task activations; control is a body of knowledge providing a system with dynamical focus of attention. Tasks are analysed by means of examples from the geophysical domain. Strong and weak points of the KBS approach are also discussed, to give an insight into achievements and current limitations: researchers in signal processing are expected to play an important rôle in future developments.

Design-concept generation for Instrument systems: A knowledge-based system approach

This paper presents further developments on a knowledge-based system (KBS) for conceptual design of instrument systems. The knowledge base of the KBS contains functional descriptions of physical components, commonly used in instrument systems. These consist of the input and output for a component, the operating signal, frequency ranges and impedance characteristics, the frequency response, the steady-state transfer function and the resolution for the component. The knowledge representation of the system is based on a hierarchical frame-based structure which has a highly modular and flexible nature. From the fundamental requirement specifications the KBS can generate a number of different combinations of physical subsystems which satisfy the requirements of an instrument designer at a power flow level of representation. An example to illustrate how the system works is finally presented.

Recent Developments in Chip Control Research and Applications

Summary During the 1990 CIRP General Assembly in Berlin a working group on Chip Control was formed within the STC -“Cutting”, with a view to evaluating and sharing the present knowledge on chip formation and chip control and promoting new research in this area. To date, about 40 researchers have made more than 60 contributions at six chip control working group meetings. Among these presentations, several contributions really provided fresh insights dealing with new and innovative research methods. The working group also provided a framework for international cooperative work, enabling exchange of information and coordinated research collaboration. The group placed a greater emphasis on the practical value of knowledge on chip control for industrial applications. This paper presents the major results of this cooperative work covering modeling of the chip formation process for chip flow; curl and breaking; means of chip breaking, including experimental as well as the knowledge-based systems approach; and a summary of future directions in research and applications.

Concurrent tool and production system design for a surface broach cutting tool: a knowledge-based systems approach

This paper presents the economics of broaching operations, as analysed within the framework of concurrent design of the broach tool and its operating conditions. The tool is specified by its cutting teeth—tooth structure, tooth pitch and the number of such teeth—which in turn determine the length of the tool. Its operation is specified by the depth of cut per tooth and by the speed of broaching. Design constraints include broaching machine capabilities and surface finish requirements. A nonlinear mathematical optimization model, which incorporates all design variables and constraints, is developed and used to analyse broach design and operation for an illustrative case.

Quality control in nerve conduction studies with coupled knowledge-based system approach.

Contemporary equipment used for nerve conduction studies is usually capable of computerized measurement of latency, amplitude, duration, and area of nerve and muscle action potentials and resulting conduction velocities. Abnormalities can be due to technical error or disease. Identification of technical error is a major element of quality control in electromyography, and artificial intelligence could be useful for this purpose. We have developed a coupled knowledge-based prototype system (QUALICON) to assess the correctness of recording and stimulating characteristics in routine conduction studies. QUALICON extracts numeric features from CMAPs or SNAPs, which are translated into symbolic form to drive a Bayesian network. The network uses high-level knowledge to infer the quality of stimulating and recording electrode placement as well as polarity and stimulus strength making recommendations as to the likely technical error when abnormal potentials are detected. A preliminary assessment shows that QUALICON performs as well as manual assessment performed by professionals.

Backcalculation of layer moduli from nondestructive pavement deflection data using the expert system approach : Nondestructive Testing of Pavements and Backcalculation of Moduli, ASTM STP 1026. Bush, A.J. and Baladi, G.Y. (Ed.) pp. 341–354. American Society for Testing and Materials. (1989)

None of the existing backcalculation procedures is guaranteed to give a reasonable, effective elastic layer modulus for every deflection basin measured. The differences of results among various backcalculation programs are magnified by the differences if distinct analysts using the same program but applying different input parameters. A more experienced analyst often obtain better results. Some of the difficulties, such as the deviation of pavement material properties from the linear elastic, isotropic, and homogenous assumptions, the uncertainty of input parameters (for example, the deflection measurement), and the limitation of basin matching techniques, may only be dealt with by informed engineering judgment. A knowledge-based expert system approach as described in this paper may be used to generate more consistent and better estimations of in situ pavement material properties by combining numerical computations with a pavement experts knowledge and judgment. A knowledge base which serves as an intelligent pre- and postprocessor to the backcalculation program is developed. The knowledge base may be revised easily as better understanding emerges.

A prototype knowledge-based expert system for the design and detailing of reinforced concrete beams

Most computer programs developed for reinforced concrete beam design either analyze a given cross section for load capacity or offer a limited solution from which the engineer must continue toward the final design. These programs are very useful in performing many tedious calculations. However, they generally do not offer assistance in areas of design that require intuitive reasoning, experiential knowledge, rules of thumb, and sound engineering judgement. This type of heuristic knowledge has been incorporated into a prototype knowledge-based expert system for reinforced concrete beam design and detailing called BeamEx. BeamEx interacts with the user to advise and establish suitable parameters required for the comprehensive design of rectangular simply supported and continuous beams. It incorporates heuristic rules drawn from the governing Canadian codes and textbooks to design alternative suitable beams which are presented in graphical form. It is shown that a knowledge-based system approach can be used effectively in engineering design by encapsulating domain expertise in a program to complement and check the experience of the users in design. Key words: knowledge-based expert system, reinforced concrete, beams, design, detailing.

Knowledge-based systems approach for offshore safety training

A number of training methods has been used within the offshore industry to impart safety knowledge and safe operating skills. In this paper a Knowledge-Based Safety Training architecture is proposed. The architecture uses both expert system and hypertext methods and is also applicable to plants other than offshore platforms. Special attention is given to the types of knowledge dealth with in safety training and how to represent this knowledge and teach it using various teaching strategies.

Optimum selection of grinding parameters using process modelling and knowledge based system approach

The paper describes the development of a computerised system which makes use of knowledge engineering and process modelling for the optimum selection of grinding parameters. Based on the theoretical analysis of experimental data on basic grinding parameters, wheel wear and specific energy, process models have been developed which are constructed in such a way that they can be dynamically modified according to user input by a rule based system. The process models accommodate such grinding parameters as grinding force Ft′, specific energy Es, temperature Tmax, wheel wear parameters etc. The first stage of the selection procedure yields specifications of an appropriate wheel and a set of nominal grinding parameters for a given grinding situation. These can be evaluated in the second stage for burn-free grinding and/or grinding with maximum grinding ratio (‘G’). A repeated use of the evaluation procedure can yield optimum grinding parameters for a desired criterion, e.g., low grinding force, good surface finish.

Optimum selection of grinding parameters using process modelling and knowledge based system approach

The paper describes the development of a computerised system which makes use of knowledge engineering and process modelling for the optimum selection of grinding parameters. Based on the theoretical analysis of experimental data on basic grinding parameters, wheel wear and specific energy, process models have been developed which are constructed in such a way that they can be dynamically modified according to user input by a rule based system. The process models accommodate such grinding parameters as grinding force Ft′, specific energy Es, temperature Tmax, wheel wear parameters etc. The first stage of the selection procedure yields specifications of an appropriate wheel and a set of nominal grinding parameters for a given grinding situation. These can be evaluated in the second stage for burn-free grinding and/or grinding with maximum grinding ratio (‘G’). A repeated use of the evaluation procedure can yield optimum grinding parameters for a desired criterion, e.g., low grinding force, good surface finish.

Weaning patients from mechanical ventilation. A knowledge-based system approach

The WEANing PROtocol ( weanpro) knowledge-based system assists respiratory therapists and nurses in weaning post-operative cardiovascular patients from mechanical ventilation in the intensive care unit. The knowledge contained in weanpro is represented by rules and is implemented in M.1® by Teknowledge, Inc. weanpro will run on any IBM®-compatible microcomputer. weanpro's performance in weaning patients in the intensive care unit was evaluated three ways: (1) a statistical comparison between the mean number of arterial blood gases required to wean patients to a T-piece with and without the use of weanpro, (2) a critique of the suggestions offered by the system by clinicians not involved in the system development, and (3) an inspection of the users acceptance of weanpro in the intensive care unit. The results of the evaluations revealed that using weanpro significantly decreases the number of arterial blood gas analyses needed to wean patients from total dependance on mechanical ventilation to independent breathing using a T-piece. In doing so, weanpro's suggestions are accurate and its use is accepted by the clinicians. Currently, weanpro is being used in the intensive care unit at the East Unit of Baptist Memorial Hospital in Memphis, Tennessee.

The Brand Manager's Assistant: A knowledge-based system approach to brand management

This paper describes the conceptualization and prototyping of the Brand Manager's Assistant, a knowledge-based system approach to the brand management function. A view of brand management is presented, followed by a conceptual framework for building marketing systems to support the brand management function. Then, the notion of object-oriented design is introduced, followed by a description of the implementation of the conceptual framework in an object-oriented programming environment. An example implementation is used to illustrate the concepts and the object-oriented design approach.

Frame-based implementation of a design knowledge-capture scheme

A knowledge-based system approach to a design knowledge capture (DKC) problem is presented. Such a system would capture not only the traditional definitive knowledge which describes a design, but also the explanatory knowledge which explains how the design decisions were made. Typical design processes were examined to determine what knowledge to retain and at what stage that knowledge is used. An outline of a DKC system was implemented using the Forms frame system. Although the problem of determining an optimum knowledge decomposition scheme is complex, a particular knowledge decomposition scheme useful to the design of magnetic bearing-based systems was imposed. It was determined that a DKC system is an extremely valuable engineering tool for complicated systems and systems with long service life.

Use of knowledge-based systems for the review of environmental impact assessments

The quality of an environmental impact assessment (EIA) and public participation can improve if the public can effectively review the EIA document. Public groups would benefit from having available specific knowledge needed to perform good reviews. A knowledge-based system (KBS) can potentially disseminate this knowledge in an inexpensive and readily usable form. Research presented in this paper evaluates the usefulness of the KBS approach by developing a research prototype capable of critiquing a central component of EIA, the evaluation of alternatives. Case studies are used to demonstrate the system's effectiveness and the quality of its interaction with nonexperts.

Knowledge-based system support for the information center's troubleshooting function

Abstract Knowledge-based systems support for the consulting/troubleshooting function within information centers was investigated. This function is defined and characteristics are identified which make it an appropriate target for knowledge-based systems support. A general architecture for a troubleshooting expert system was developed and its elements are described. A case is described which illustrates the application of an expert system in the provision of troubleshooting service by the information center of a major state university. The results illustrate the impact that such a knowledge-based system approach can have in the troubleshooting area, especially in relation to the efficient utilization of information center personnel.

A knowledge-based approach to support the generation of construction schedules

Computerized support of the task of generating construction schedules is currently limited to provide a network representation of activities manually defined and sequenced. This definition of the activities, their characteristics (resource allocation, duration, etc.), and their sequencing demands substantial expertise and is time consuming. This research work attempts to provide intelligent support to the construction scheduler. A Knowledge-Based Systems approach is used, which involves three major phases: (1) knowledge acquisition from experts; (2) formalization of the elicited knowledge; and (3) production of a prototype Knowledge-Based System. An extensive program of interaction with industry expert schedulers is described here. The status of the computer implementation work is discussed as well.

A knowledge-based system approach to determine durations of site activities

A knowledge-based system, Ratu-aj, has been developed for determining durations of site activities. It is based on the Finnish system of production files, Ratu-files, developed for new build construction work. Ratu-aj is structured hierarchically and combines lower level work task durations into a higher level activity duration. Standardized computer screen forms and graphical symbols make Ratu-aj easier and quicker to use than under manual conditions. The current version is employed in estimating the durations of large shuttering panel tasks.