“EXPERT SYSTEM” DESIGN PROJETOS NA ÀREA: Development of an Expert System as a Design Tool www.catd.iastate.edu/public/Coop_Projects/ development_of_an_expert_system_.htm Company Size: 6,000 employees Project Costs: $37,300 Timeline: March, 1997 – August, 1998 Project Description: The purpose of this project was to pursue the development of a knowledge-based expert system that would enhance the overall design process typically undertaken by the company’s engineers. An expert system would compliment the design process by providing a tool that would allow designers access to relevant historical data and to introduce reason-based filtering of the database. Researchers within the Aerospace Engineering and Engineering Mechanics Department concentrated on a specific area of work within the company. Using this area as a case study, a prototype computer software program was developed. The prototype system can be used, tested, and modified. This expert system was designed to be expandable into new areas of research and development with the advantage of saving time, preserving knowledge, and guiding design projects Event Monitoring Enabling Responses to Anomalous Live Disturbances SRI – INTERNATIONAL SYSTEM DESIGN LABORATORY 10/23/2001 http://www.sdl.sri.com/programs/intrusion/ EMERALD Overview The EMERALD (Event Monitoring Enabling Responses to Anomalous Live Disturbances) environment is a distributed scalable tool suite for tracking malicious activity through and across large networks. EMERALD introduces a highly distributed, building-block approach to network surveillance, attack isolation, and automated response. It combines models from research in distributed high-volume event correlation methodologies with over a decade of intrusion detection research and engineering experience. The approach is novel in its use of highly distributed, independently tunable, surveillance and response monitors that are deployable polymorphically at various abstract layers in a large network. These monitors contribute to a streamlined event-analysis system that combines signature analysis with probabilistic inference to provide localized realtime protection of the most widely used network services on the Internet. The EMERALD project represents a comprehensive attempt to develop an architecture that inherits well-developed analytical techniques for detecting intrusions, and casts them in a framework that is highly reusable, interoperable, and scalable in large network infrastructures. EMERALD presently includes several eXpert components that use PBEST rule based inference. These analyze network and host based data and comprise the most complete knowledge base in the field. EMERALD also includes eBayes components that use Bayesian inference. SPEED® Sizing and Preliminary Design of Electric Motors. http://www.cedrat.com/software/speed/speed.htm SPEED software is backed by over decade of experience within the Speed Laboratory of the University of Glasgow. SPEED software is PC based CAE software for the design and performance calculation of electric motors including parameters of power electronic drives used with these motors. Motor design with SPEED software is interactive and fast, as SPEED is based on traditional and validated analytical theories. Its main use is : Sizing and preliminary design Studying the performance and quickly assessing the effects of parameters change. PC-IMD PC-BDC PC-SRD PC-DCM AC induction motor Polyphase and Split-phase Brushless PM motor Switched reluctance motor DC commutator motor An expert system eases rotor design The KBE system can be used to design turbine-engine rotor disks The first objective was to create one model of the system, then allow different engineering specialties to view that model through their own analytical filters, keeping all specialties updated as the design evolves. Component interactions can frequently lead to mismatched components, so the KBE system gives as much importance to system integration as to component design. The system requires certain components to be sized ahead of others, rather than all proceeding individually, to avoid downstream surprises and risk to the program. A KBE system pays for itself if one major redesign can be avoided. The above was adapted from an article by John Marra, a project engineer at Pratt & Whitney’ s Design System Technology Group in West Palm Beach, Fla. The full text may be found in the April 1997 issue of Mechanical Engineering magazine. © 1997 ASME Bus-Expert-System INSTITUTE OF MECHATRONICS: AT THE CHEMNITZ UNIVERSITY OF ECHNOLOGY http://www.tu-chemnitz.de/ifm/english/projekte/ep_bus.htm Project Title: Bus-Expert-System for Dynamic Simulation, Design and Quality Control Supported by: EU Programme COPERNICUS Working Period: 04/95 - 06/98 Summary of the Project The aim of the research was the development of a powerful simulation tool especially for producers of busses, their suppliers and research institutes connected with them. A computerprogram-system for dynamics simulation and computations developed by using the newest methods of the applied mathematics and mechanics is capable of an effective designing and construction of busses, of the evaluation of materials, of the investigation of dynamic behaviour, of the quality control and the improvement of reliability of busses. Extensive experimental investigations, measurements and checks demonstrate the quality of the software and extend the applicability of the system. The enterprises SKODA Ostrov (Czech Republic), IKARUS (Hungary) and ROCAR (Romania) are integrated in the joint research as first users. It is intended a further distribution of the system in other European countries. The most important tasks and results of the joint research are : Development of a powerful bus-modell with wheels, tires, axis, engine, steering system, passenger cab, pantograph (trolleybus) and passengers; computersimulations using this model for all essential driving situations and conditions including overturning of bus and crashtest; extensive experimental investigations on testbeds and while driving of the bus for validation and optimization; computations and experimental investigations of stresses, properties of materials (fatigue) and fatigue life estimation; development of a software-system having an own graphical interface. Improved Well Performance Research Program Sandstone Acidizing Design Integrated and Evaluation September 8, 1997 http://www.pe.utexas.edu/CPGE/IWP/acid2.html OBJECTIVES OF THE PROJECT In recent years, we have developed a two-acid, three-mineral sandstone acidizing simulation model which predicts bottomhole pressure and skin factor change as a function of injected acid volume. We have also developed a preliminary expert system model which provides recommendations for acid type and volume, additives, preflush, and postflush for sandstone acidizing design. The current simulation model constrains the injection by either fixed bottomhole pressure or fixed total injection rate but does not consider the limitation of tubing flow on acid injection. We propose in this project to add a wellbore model to the simulation model, to upgrade the expert system model, and then to integrate the expert system model and the simulation models into a practical package to optimize sandstone acidizing design. The design model then will be attached to a real- time monitoring model for acidizing treatments. Both the predicted skin factor from the model and the estimated skin factor from the measured pressure and injection rate will be plotted simultaneously during the treatment to allow the operator to evaluate and supervise the treatment closely. The final product of the project will be a PC-based Windows program. It will feature Windows graphics and a summarized report document and will be easily usable in the field. Transonic Airfoil Design with Expert Systems Robert Zores http://www.sm.go.dlr.de/~bob/Docu-extern/AIAA/AIAA95.html#HDR0 Abstract The combination of new technologies with classical methods for designing transonic airfoils is described in this paper. While designing supercritical airfoils the skilled aerodynamicist is usually completely aware of the laws and criteria that guide the designing process. An expert system can help inexperienced engineers reduce the time of learning and exploring new generated airfoils. It speeds up the process of generating, calculating and analyzing the airfoil. Proven programs, such as inverse CFD solvers, beside new technologies from computer science, can be used to build new tools for aerodynamic design on workstations. A selection of improved workstation tools to design transonic airfoils is described. The capability of the expert system is shown by two examples. The first example raises a more academic question on airfoils which can be concave on their upper side. The second example shows how an expert system can be used to support wind tunnel tests. October 29-31, 1996, Seoul, Korea. DESIGN EXPERT SYSTEM FOR AUTO-ROUTING OF SHIP PIPES ABSTRACT http://icad.kaist.ac.kr/~msh/ARS/cad-ars.htm Finding the optimum route of ship pipes is a complicated and time-consuming process. Experience of designers is the main tool in this process. To reduce design manhours and human errors a design expert system shell and a geometric modeling kernel is integrated to automate the design process. A framework of the intelligent CAD system for pipe auto-routing is suggested. The CADDS 5 of Computervision is used as the overall CAD environment, the Nexpert Object of Neuron Data is used as the expert system shell, and the CADDS 5 ISSM is used to build user interface through which geometric models of pipes are created and modified. Existing algorithms for routing problems have been analyzed. Most of them are to solve 2-D circuit routing problems. Design of ship piping system, especially within the engine room, is a complicated, large scale 3-D routing problem. Methods of expert systems have been implemented to find routes of ship pipes on the main deck of a bulk carrier. Key Words : Pipe auto-routing , Expert system , Geometric modeling , Intelligent CAD, Ship design ActivEX Expert System for Design and Optimization of Active Material Based Devices: Dielectric Elastomers and Shape Memory Alloys http://www.esit.com/defense/activex.html The latest implementation of ActivEX Expert System is aimed at the broad market of biomedical devices that are designed using last Smart Electroactive Materials (e.g., shape memory alloys and electroactive polymers). The ActivEX package has been originally conceived to synthesize and evaluate design alternatives for ESI's ongoing R&D project dedicated primarily to the development of a new-generation ElectroStrictive Anti-G Suit based on Smart Active Materials. A broad variety of biomedical, industrial, and consumer devices (including such up-and-coming gimmicks as electroactive fabric, and self-fitting clothes and bras) can be designed based on Smart Materials using ActivEX. Some examples that are commercialized by a number of lead players on the US field of medical devices include electroactive stents for supporting a blood vessel, vasooclusion coils, variable stiffness guidewires, steerable catheters. ActivEX effectively includes a Computerized Handbook of Standard Application Modules (SAM), such as human body parts with their associated ActivEX-based models reflecting important design properties (e.g., a body part compliance) and Smart Active Material based Standard Automation Modules (SAM2) together with ActivEX-based design rules and LabVIEW-generated experimental data. GUI and working environment of ActivEX has been developed in C++ within AutoCAD Mechanical Desktop. The Automated Test Planning System: An Expert System Tool for T&E Oversight http://www.stsc.hill.af.mil/crosstalk/1994/nov/automate.asp M. Scott Roth, Science Applications International Corporation (Editor's Note: This article is a reprint from an article published in the September-October 1994 issue of the Program Manager, a publication of the Defense Systems Management College (DSMC). The article appears with the permission of DSMC and the author.) Abstract The Automated Test Planning System (ATPS) is a set of expert system-based tools designed to aid the Office of the Under Secretary of Defense for Acquisition and Technology/Director of Test and Evaluation (OUSD(A&T)/DT&E) staff and DoD Service component staffs in the test and evaluation oversight process. The TEMP Review Module of the ATPS provides an intelligent system to aid in the review of Test and Evaluation Master Plans (TEMPs). The Test and Evaluation Program Risk Assessment Module aids in the harmonization of key acquisition documents and highlights risk indicators of a T&E program. The TEMP Build Module, currently under development, will assist in the construction of a TEMP by providing structure to the development process and by providing on-line guidance to the user. The ATPS combines highly effective technologies (such as expert systems, hypertext, and databases) into a seamless environment and presents these capabilities in a user-friendly manner. The ATPS currently operates in the Microsoft Windows and Apple Macintosh environments. http://www.ee.ed.ac.uk/ESG/power-electronics.html Expert System Design of Switched Mode Power Supplies Amarnath Reddy (in collaboration with Minebea Electronics (UK) Ltd.) Supervisors: Dr. DE Macpherson, Prof. HW Whittington This project investigated using an Expert System to design SMPS. The expert system was linked to databases of common circuit topologies, semiconductor components, magnetic cores etc. The design included component selection, design of wound components, error amplifer compensation, and design for EMC. Design verification was carried out by linking the expert system to SPICE, allowing full circuit simulation to be carried out. - Associate Technical Director Http://easydiagnosis.com/software.html David Schachter Expert System Programs Symptoms, signs, and conditions For over 100 years medical diagnosis has been largely based on invoking lists of disease possibilities. Thus, medical texts and references, such as the Merck Manual® are organized by disease, or conditions. Medical teaching has generally been based on the same formulation, emphasizing the study of disease, not the study of complaints, i.e. Given this disease, do the patient's complaints ( symptoms) or findings (signs) match the diagnosis? This is backwards, of course, and most physicians know this instinctively. Since patients do not present with diagnoses, the logical question should be: Given this symptom or sign, what are the most likely conditions or diagnoses, in descending order of their probabilities. Because of the way clinical data has been collected anecdotally over the past century, with emphasis on disease occurrence rather than the prevalence of signs and symptoms, this presents significant difficulties. EasyDiagnosis, well aware of the limitations of this approach, has attempted with these programs to present this concept in a coherent, simplified format. The vast majority of patients present with one major symptom or sign, and in a significant majority of these, the top three or four diagnostic probabilities usually add up to over 70%-80%. When there is high suspicion of a specific condition, our method can be used to estimate roughly the probability of its occurrence in a given setting. Intelligent Systems Knowledge Engineering/Artificial Intelligence http://www.acrc.unisa.edu.au/is/keai/ The traditional approach to developing an ``expert system'' has been to interview a domain specialist until all the rules (knowledge) this person applies in practice are represented for use by the computer. This approach has many limitations in practicability and effectiveness. We focus on artificial intelligence techniques that achieve intelligent-seeming behaviour in control, classification and diagnosis without relying on this traditional approach to knowledge engineering. A number of projects in this topic are currently in progress: Advanced Logic Controllers, investigating a fuzzy logic control of electronic wheelchairs which handles erratic and uncertain input caused by users' hand tremors. AI Techniques in Distributed Processing, studying techniques such as artificial neural networks and internal arithmetic for distributed and parallel computing. Automatic Generation of a Neural Network Architecture using Evolutionary Computation, studying techniques for directly encoding a neural network in the genetic tree structure in genetic programming environments. Computer-aided Pronunciation System (CAPS), developing computer-aided pronunciation systems to provide meaningful feedback on pronunciation to students of languages in a stand-alone language learning system. Development of New Techniques for Enhanced Resolution for SAR Systems in Intelligent Environment, improving the information provided by radar systems. Diagnosing Engineering Systems in Intelligent Environment, studyiong system diagnosis in a knowledge-based environment. An Efficient SAR Algorithm in Intelligent Environment, developing an efficient algorithm which transforms SAR data in one operation. Entropy Functions for Focussing ISAR Images in Intelligent Environment, developing a technique which provides an approximate polar reformatting of an ISAR target which is independent of the rate of rotation. Evaluation of Artificial Neural Networks, developing new techniques for the cross-validation of artificial neural network classifiers. Expert System Construction by Reasoning from Cases, applying decision tree induction to the development of case-based reasoning systems. Exploiting Conceptual Graphs for Knowledge Reuse in Design, developing a pattern-directed inference system over conceptual structure terms. DEIS - Università di Bologna - L I A - Laboratorio d'Informatica Avanzata http://www-lia.deis.unibo.it/Research/Areas/knowrep.html#ExpertSysDes General Description The aim of this research theme is to show how models, techniques and tools for building expert systems can help to face real problems. Jointly with SASIB S.p.A. an expert system based on Prolog and metainterpretation has been implemented in order to automatically design railway signalling systems. Moreover, optimization techniques have been applied, mainly based on partial evaluation. Meta-interpretation has also been adopted to implement a simulator for the designed system. Further, the management and fault diagnosis in big and medium-size stations has been faced by using a tool based on a parallel blackboard model. Recently, the same problem has been solved by using KEE as implementation tool and its Assumption-based Truth Maintenance System, and Constraint Logic Programming. Intelligent Tutoring Systems (ITS) have the purpose of transferring domain knowledge and experience to a user (student). ITSs evolved from the earlier Computer Aided Instruction (CAI) systems; they represent knowledge declaratively and may adapt their behaviour to the student's characteristics. The problems of modelling the domain, the students' characteristics and the communication process are thus extremely relevant to the design of an ITS. The research applied Description Logics to the design of an ITS prototype for the instruction of staff of a metal and mechanics manufacturing company. The focus has been on the student's model and classification, showing how Description Logics can be a substantially sound support to the solution of this kind of problems. An ITS architecture has been proposed, based on the notion of student stereotype and knowledge granules. The student and granule representation models have been studied and the algorithms for the choice of the knowledge granules to be taught to a student have been developed. Participants Sonia Bergamaschi Maurelio Boari Evelina Lamma Paola Mello Claudio Sartori Funded by 1986-94: SASIB S.p.A. Bologna, 1992-94: G.D. S.p.A. Bologna , MURST 60% - Methods and Tools for Building Expert Systems LIMEX: An Integrated Expert System With Multimedia http://potato.claes.sci.eg/claes/limex/limex.html Introduction and highlights on each function of LIMEX LIMEX is an integrated expert system with multimedia that has been developed to assist lime growers and extension agents in the cultivation of lime for the purpose of improving their yield. The scope of LIMEX expert system includes: assessment , irrigation , fertilization , and pest control . A preliminary evaluation process, involving both domain experts and end users, have been to study their needs and the feasibility of building LIMEX. The study showed that there is a great need to an expert system because of the complexity of operations which effects the tree yield in different seasons. LIMEX is being promoted and supported by CLAES which sees to it that knowledge is updated on a regular basis.The expert system was augmented with multimedia capabilities as enhancing an expert system by the integration of text, image, sound, video, and data, allows for a good feedback from users, assists in better understanding of the system, and allows for more flexibility in the interactive use of the system. The purpose of integrating the LIMEX expert system with multimedia was to develop functional modules capable of supplying end users with adequate information and advice to solve their problems. LIMEX is divided into subsystems, each of which is concerned with a specific problem. Figure 1 illustrates the hierarchy of LIMEX subsystems and functional modules. Assessment , Irrigation , Fertilization , Pest Control ,