Knowledge-based Module Research Articles

An integrated artificial neural network-computer simulation for optimization of complex tandem queue systems

This paper presents an integrated artificial neural network-computer simulation (ANNSim) for optimization of G/G/K queue systems. The ANNSim is a computer program capable of improving its performance by referring to production constraints, system's limitations and desired targets. It is a goal oriented, flexible and integrated approach and produces the optimum solution by utilizing Multi Layer Perceptron (MLP) neural networks. The properties and modules of the prescribed intelligent ANNSim are: (1) parametric modeling, (2) flexibility module, (3) integrated modeling, (4) knowledge-base module, (5) integrated database and (6) learning module. The integrated ANNSim is applied to 30 distinct tandem G/G/K queue systems. Furthermore, its superiority over conventional simulation approach is shown in two dimensions which are average run time and maximum number of required iterations (scenarios).

The DSCP-CA

The purpose of this study was to develop a decision support system using fuzzy logic that would support nurses' decisions about cancer pain management, especially for ethnic minority cancer patients. The study had two phases: (1) data collection and (2) development of the decision support computer program. In the data collection phase, an Internet survey of 428 cancer patients and four ethnic-specific online forums (about 30 participants per forum) were conducted to gather data on the cancer pain experience of the four major ethnic groups in the United States. The development phase included two components: (1) development of three modules, including a knowledge base module, a decision module, and a self-adaptation module; and (2) a 3-month evaluation of the decision support computer program by oncology nurses and subsequent incorporation of their feedback into the program. The Internet survey and online forum data were processed into fuzzy and crisp data sets, and ethnic-specific algorithms for the decision module were developed. Using the self-adaptation module, the decision support computer program was further refined as additional data were processed. Then, the decision support computer program was further developed by adding additional components suggested by the oncology nurses.

Intelligent decision support for structural design analysis

Many comprehensive composite computer-aided design systems with special analysis tools have been developed in order to make the product development process more effective. However, these computer-aided tools still concentrate mainly on the numerical aspects of analysis whilst, in most of cases, they fail to provide any kind of expert advice during the decision-making process. It is obvious that the existing computer tools in this field need to be extended in order to reach a higher level of performance by adding some intelligent behaviour. This paper discusses a framework for intelligent decision support for structural design analysis using a finite element method. The intelligent environment presented is composed of several knowledge-based modules in order to address some major bottlenecks within the analysis-based design improvement process. The prototypes of four intelligent modules are presented within this context. These are: an intelligent module for supporting some initial decisions within the design analysis process, an intelligent module for finite elements’ selection, an intelligent module for finite element mesh design, and an intelligent module for supporting the results’ interpretation process. According to the feedbacks of those experts who participated in the evaluation process, these prototypes could be applied as useful supporting tools for inexperienced designers in design practice, as shown in certain examples presented in this paper.

An automated design system for progressive die

This paper describes an automated design system developed for design of progressive die. The proposed system is organized in 27 knowledge-base modules. The production rule based knowledge-base system (KBS) approach is utilized for constructing the system modules. Modules are user interactive and designed to be loaded into the prompt area of AutoCAD. The system is capable of automating all major activities of design of progressive die such as checking the design features of sheet metal parts, design of strip-layout, selection of progressive die components, modeling of die components and die assembly, and selection of materials for progressive die components. The system is implemented on a PC having AutoCAD software and therefore its low cost of implementation makes it affordable by small and medium size enterprises.

Development of an Expert Engineering Module for Determination of Ultrasonic Probe Position on the Weld Joint of Plate

Ultrasound based inspection techniques are used extensively throughout industry for detection of flaws in engineering materials. The range and variety of imperfections encountered is large and critical assessment of location, size, orientation and type is often difficult. The research presented in this paper has resulted in a knowledge-based engineering module, which was developed software for determining ultrasonic probe location precisely and detecting of the welded sections completely. The criteria for diagnosing these positions, which have been classified into different groups, included in type of material, thickness, geometry of welded section, couplant material, angle probes and etc. The expert system can display the limits that probe must be moved which caused an inspection welded section thoroughly. This study was conducted using three samples including different materials with V groove welded sections for verifying this expert system. Finally this module is very suitable in Q.C. labs and different industries for detection of flaws.

Developing a group decision support system based on fuzzy information axiom

Information axiom, one of two axioms of axiomatic design methodology which is proposed to improve a design, is used to select the best design among proposed designs. In the literature, there are a lot of studies related to using of information axiom for the solution of decision making problems. Moreover, applications of information axiom have been increasing day by day. However, calculation procedure of information axiom is not only incommodious but also difficult for decision makers. In this paper, a decision support system (DSS) based on fuzzy information axiom (FIA) is developed in order to make this decision procedure easy. The developed system consists of a knowledge base module including facts and rules, inference engine module including FIA and aggregation method, and a user interface module including entrance windows. The main aim of this study is to present a DSS tool to help the decision makers to solve their decision problems by modifying data-base of the program. In this paper, an application procedure will be presented based on the optimal selection of location for emergency service to illustrate the implementation procedure of the proposed model.

모델러와 CAE 를 활용한 최적 금형 방안 설계 기술 개발

In the casting industry field, a lot of researches have been carried out to maximize soundness and productivity and improve quality of mass produced products. The researches on the optimum mould design by using computer simulation have achieved a lot of accomplishments in the die-casting process especially. Their results are utilized to mass-produce sound products in their field. But there are some difficulties to use the simulation in order to make the design of product and mould. In general, repetitive calculations of computer simulation require much analyzing time in order to make sound design of them. Another difficulty is that only expertise technicians can analyze results of the simulation. In order to solve such a problem, we are building mould design optimization system which consists of knowledge-based module, simulation, optimization module and interactive modeler(it is based on the commercial modeler). By using and interacting between these modules of that system, we can easily make an optimum design of mould and construct the database. This study introduce the general process of the system and connection method between the modules is mentioned above and some examples to apply it to real product.

An Internet-based intelligent design system for injection moulds

The rapid growth of Internet and information technologies in recent years provides a solution to support and facilitate collaborative product developments among different geographically distributed enterprises. An effective and feasible tool to aid the collaborative development of injection moulds can be realized by developing an Internet-based mould design system as one of the modules of a collaborative product development system. This paper presents a prototype Internet-based intelligent design system for injection moulds. The architecture of the system consists of an interactive KB mould design system embedded in an Internet environment. A Java-enabled solution together with artificial intelligence techniques is employed to develop such a networked interactive CAD system. In this system, the computational module, the knowledge base module and the graphic module for generating mould features are integrated within an interactive CAD-based framework. The knowledge base of the system would be accessed by mould designers through interactive programs so that their own intelligence and experience could also be incorporated with the total mould design. The approach adopted both speeds up the design process and facilitates design standardization which in turn increases the speed of mould manufacture. A practical case study is presented to illustrate the operations of the Internet-based mould design system.

A short note on an intelligent system for selection of materials for progressive die components

Selection of materials for die components is an important activity during progressive die design in stamping industries. This paper presents an intelligent system for selection of materials for progressive die components. The proposed system SMPDC comprises of two knowledge base modules, namely DIEMAT and SELHRD. The module DIEMAT is designed for selection of materials for both active and inactive components of progressive die. The module SELHRD is developed for determination of hardness range of materials for active components of progressive die. Knowledge for both the modules of the proposed system is acquired, analyzed, tabulated and incorporated into a set of production rules of IF-THEN variety. The system is coded in the AutoLISP language and loaded into the prompt area of AutoCAD. The system is designed to interact with the user through the user interface. The usefulness of the proposed system is demonstrated through a sample run using an example of an industrial component. The knowledge base of the system can be modified depending upon the availability of new materials and advancement in technology.

Trends and Developments in Intelligent Computer Aided Design of Progressive Dies

This paper discusses the trends and developments in intelligent CAD of progressive die. The research efforts reviewed reveal the growing importance and relevancy of particular AI techniques for automation of progressive die design activities. Based on the knowledge base approach of AI, an intelligent CAD framework is developed and presented in this paper. The proposed intelligent CAD framework comprises different knowledge base modules for automating progressive die design process. The procedure for the development of various knowledge base modules of the proposed intelligent CAD system is described through module CCKBS which has been designed for assessing manufacturability of sheet metal parts. An illustrative example has been included for demonstrating the usefulness of the developed module CCKBS. The system framework is flexible enough to accommodate new acquired knowledge for each module. The modules of the system framework are implementable on a PC and thus can be affordable by small and medium size enterprises.

MACHINE–LEARNING ASSISTED KNOWLEDGE ACQUISITION TO FILTER LACTATION CURVE DATA

Machine learning was employed to develop knowledgebased modules to automate the filtering of testday datafor groupaverage lactation curve analysis. The classification task consisted of identifying an individual cows milkyieldtest as being either outlier or nonoutlier. Experiments were carried out to explore the importance of deriving predictiveattributes and choice of machinelearning technique. Data consisted of 1080 milk yield tests, of which 108 had been classifiedas outliers by a dairy nutrition specialist and 972 cases had been classified as nonoutliers. Two different approaches tomachine learning were applied to these data: decisiontree induction and naveBayes classification. Performance of theclassifiers was estimated through tenfold crossvalidation, while relative operating characteristic curves were used tovisualize the achieved tradeoff between sensitivity and specificity. Use of an initial set of derived attributes significantlyimproved the performance compared to limiting attributes to those available to the domain specialist. However, adding evenmore complex attributes did not necessarily improve the classification performance. Overall, the naveBayes approachshowed significantly better performance than decisiontree induction. For each machinelearning approach, three finalclassifiers, associated with a low, medium, and high filtering intensity, were generated from the 1080 cases. The expected truepositive rate varied from 41% to 78% for false positive rates between 1.1% and 4.6%. However, due to the low prevalenceof outlier tests, this performance was associated with a large number of false positives. The domain specialist considered thefinal classifiers of both approaches as plausible, but found the decision trees easier to understand than the naveBayesclassifiers. Machine learning was considered to be a promising approach to assist knowledge acquisition.

A computer-based intelligent system for automatic tool selection

This paper presents a computer-based intelligent system for the automatic selection of cutting processes and tools within a concurrent engineering environment. The main objective of the research work was to develop a procedure for the selection of cutting tools, develop a dynamic programming-based system that utilises mathematical modules and heuristic data to determine and calculate cutting parameters and total component cost. This will help the designers and manufacturing planners to select an optimal set of cutting tools and cutting conditions for different material properties and to give users alternatives on how to reduce cost and time. The system comprised of several modules; the knowledge acquisition module, the knowledge base module, the inference engine, the user interface, and the database. The developed system, which was designed to cover different component shapes including cylindrical, prismatic, different types of machining techniques, conventional and less conventional, has the potential to deal with complex products that are made up of multiple components. The system is capable of selecting cutting tools. It calculates cutting conditions and estimates component cost, based on the properties of the work piece material and features attributes, which include surface finish and tolerances, as well as using a number of production criteria such as material removal rate, tool life, machining time, and cost.

Concurrent Product and Process Design in Hot Forging

The long and costly lead times necessary in developing new forged products motivate the application of concurrent engineer ing principles and knowledge-based computer software systems in hot forging. Design engineers can benefit significantly from such sys- ' tems, particularly if they incorporate both product and manufacturability knowledge. An Automated Concurrent Engineering Software (ACES) for part design and manufacturing has been developed at the Design and Manufacturing Institute (DMI) at Stevens Institute of Technology. This system combines deterministic and empirical knowledge about a variety of aspects of product design and manufactur ing, and thus, it provides potent reasoning and decision-making power that enables the engineer to reduce drastically the time between product conception and final production. This paper reports on the development of the prototype of a knowledge-based module for hot forging processes.

An Interactive Knowledge-Based CAD System For Mould Design in Injection Moulding Processes

This paper presents a practical prototype knowledge-based system, called IKMOULD, for mould design in the injection moulding process. It attempts to tackle the problem in a practical and integrative way, unlike the stand-alone and mathematical programs which have been developed in the past to solve only a part of the problem. A total quantitative and structured approach is not feasible in dealing with the complex and multirelated design problems generally involved in mould design. In this system, the computational module, the knowledge-based module and the graphic module for generating mould features are integrated within an interactive CAD-based framework. The knowledge base of the system can be accessed by mould designers through interactive programs so that their own intelligence and experience can also be incorporated with the total mould design. The approach adopted both speeds up the design process and facilitates design standardisation which in turn increases the speed of mould manufacture.

A WWW-based integrated product development platform for sheet metal parts intelligent concurrent design and manufacturing

This paper presents a WWW (World Wide Web) based integrated product development platform for intelligent concurrent design and manufacturing of sheet metal parts. In order to achieve this platform, several major modules are discussed. These modules mainly include the structure of the WWW-based integrated product development platform, an information integration framework, a RTCAPP (real time computer aided process planning) module, a customer interface module, and a design/manufacturing knowledge-based module for supporting product design and manufacturing. This paper gives the structure of the information integration framework for concurrent design and manufacturing of sheet metal parts. An information integration framework, called 'step structure information framework' is proposed. The principles, called 'zero thickness and zero bend radius' are put forward, which can be used to abstract the geometric entities of sheet metal parts in order to facilitate product design and modelling. Finally, case studies are given.

Interactive multicriteria procedure exploiting a knowledge-based module to select electricity production alternatives: The CASTART system

For some decisional problems, it is difficult to have a set of potential alternatives a priori. This is true, for instance, when the object of the decision is dependent on an external environment, as in the case of disturbance situations in industrial processes. Computer assistance must support, first, the elaboration of a potential alternatives set and, second, the decision-making activity which will determine the decision elements leading to recovery from the disturbance. This paper deals with such a situation encountered in an electricity production context. An interactive multicriteria procedure exploiting a knowledge-based module to select electricity production alternatives is presented. Section 1 introduces the electricity production-consumption framework and identifies the Spinning Reserve Mobilization (SRM) function. Next, our contribution to the systematization and the experimentation of a decision-making process which supports the dispatcher confronted with an SRM situation is described. The decision-making process includes an intelligent module and a decisional procedure. The module and the procedure, and their interconnections, are developed in Section 2. Section 3 describes the DSS called CASTART which implements this process. An overview of DSS functionalities is given, as well as the results obtained by a dispatcher working in cooperation with the system, in the case of a simulated production incident.

A knowledge-based system for real-time validation of calibrations and measurements

In this contribution, a knowledge-based system is presented which is able to identify faults during measurements and calibrations of a flow injection analysis (FIA) system. The knowledge-based system is a subtask of the automation system CAFCA, which can be used for the automation of flow systems, and runs on MS-DOS-PCs. The knowledge-based system consists of a numerical module as well as a knowledge-based module, where the knowledge about the process analyser is implemented using rules. Using these rules faults will be identified fast and reliable. If a fault has been detected, the operator is informed automatically. Furthermore, minor faults can even be corrected by the knowledge-based system. The application of the system shows that it is an efficient support for the operator.

KDPAG expert system applied to materials design and manufacture

A multi-dimensional space, spanned by dimensionless numbers describing the equilibrium states or the macroscopic chemical kinetics of complicated chemical reaction systems, has been used for data processing and for the modeling of the processes needed to prepare industrial materials. It was found that there are four types of topological relationships between the optimization zone and the operation zone in multi-dimensional spaces. Pattern recognition, artificial neural networks (ANN) and genetic algorithms (GA) are used to build models and search for an optimization strategy. An integrated expert system called knowledge base, data base, pattern recognition, artificial neural network, genetic algorithm (KDPAG), consisting of a knowledge base, a database, and pattern-recognition, ANN, and GA modules has been built. Applications of this system give good results, as has been observed in several factories for the importance of the production of alloy steel, synthetic rubber, ceramic materials production and materials design of composite materials, high T c superconductors, and ceramic semiconductors.

Pragmatic user model implementation in an intelligent help system

We present an intelligent help system for the Unix operating system called Aran. Aran is a passive, knowledge‐based help system whose main goal is to help users deal with problems related to Unix operation. Aran gives the user access to different kinds of information, and it is designed to help the user accomplish a given task, while, at the same time, expanding the user's knowledge of the operating system. In order to accomplish its job Aran is endowed with an explicit representation of both domain specific knowledge and a pragmatic user model (UM) that takes into account the individual's characteristics. In Aran, user modelling is dynamic and based on stereotypes. The UM is implemented with the same representation environment, Loom, used in the rest of the knowledge base (KB) modules. This environment is based on a description logics formalism that facilitates the construction and maintenance of all the modules.

Knowledge-based supervisory control system of a fish processing workcell; Part I: system development

A laboratory workcell has been developed for fish cutting. The workcell, which uses a variety of sensors, actuators, and hardware for component interfacing and control, operates with the help of a dedicated supervisory control system. A layered architecture has been used for the system. It has several knowledge-based modules for carrying out tasks such as workcell monitoring, controller tuning, workcell conditioning, and product quality assessment. The paper describes the workcell and the development of the supervisory control system. Development of the knowledge-based modules of the system, which employ fuzzy logic for knowledge representation and decision making, is described.