Electronic commerce and the Internet have created demand for automated systems that can make complex decisions utilizing information from multiple sources. Because the information is uncertain, dynamic, distributed, and heterogeneous in nature, these systems require a great diversity of intelligent techniques including expert systems, fuzzy logic, neural networks, and genetic algorithms. However, in complex decision making, many different components or sub-tasks are involved, each of which requires different types of processing. Thus multiple such techniques are required resulting in systems called hybrid intelligent systems. That is, hybrid solutions are crucial for complex problem solving and decision making. There is a growing demand for these systems in many areas including financial investment planning, engineering design, medical diagnosis, and cognitive simulation. However, the design and development of these systems is difficult because they have a large number of parts or components that have many interactions. From a multi-agent perspective, agents in multi-agent systems (MAS) are autonomous and can engage in flexible, high-level interactions. MASs are good at complex, dynamic interactions. Thus a multi-agent perspective is suitable for modeling, design, and construction of hybrid intelligent systems. The aim of this thesis is to develop an agent-based framework for constructing hybrid intelligent systems which are mainly used for complex problem solving and decision making. Existing software development techniques (typically, object-oriented) are inadequate for modeling agent-based hybrid intelligent systems. There is a fundamental mismatch between the concepts used by object-oriented developers and the agent-oriented view. Although there are some agent-oriented methodologies such as the Gaia methodology, there is still no specifically tailored methodology available for analyzing and designing agent-based hybrid intelligent systems. To this end, a methodology is proposed, which is specifically tailored to the analysis and design of agent-based hybrid intelligent systems. The methodology consists of six models - role model, interaction model, agent model, skill model, knowledge model, and organizational model. This methodology differs from other agent-oriented methodologies in its skill and knowledge models. As good decisions and problem solutions are mainly based on adequate information, rich knowledge, and appropriate skills to use knowledge and information, these two models are of paramount importance in modeling complex problem solving and decision making. Follow the methodology, an agent-based framework for hybrid intelligent system construction used in complex problem solving and decision making was developed. The framework has several crucial characteristics that differentiate this research from others. Four important issues relating to the framework are also investigated. These cover the building of an ontology for financial investment, matchmaking in middle agents, reasoning in problem solving and decision making, and decision aggregation in MASs. The thesis demonstrates how to build a domain-specific ontology and how to access it in a MAS by building a financial ontology. It is argued that the practical performance of service provider agents has a significant impact on the matchmaking outcomes of middle agents. It is proposed to consider service provider agents' track records in matchmaking. A way to provide initial values for the track records of service provider agents is also suggested. The concept of ‘reasoning with multimedia information’ is introduced, and reasoning with still image information using symbolic projection theory is proposed. How to choose suitable aggregation operations is demonstrated through financial investment application and three approaches are proposed - the stationary agent approach, the token-passing approach, and the mobile agent approach to implementing decision aggregation in MASs. Based on the framework, a prototype was built and applied to financial investment planning. This prototype consists of one serving agent, one interface agent, one decision aggregation agent, one planning agent, four decision making agents, and five service provider agents. Experiments were conducted on the prototype. The experimental results show the framework is flexible, robust, and fully workable. All agents derived from the methodology exhibit their behaviors correctly as specified.

Transfer of information between CAD models and downstream manufacturing process planning software typically involves redundant user interaction. Many existing tools are process-centric and unsuited for selection of a "best process" in the context of existing concurrent engineering design tools. A computer based Feature-Recognition (FR) process is developed to extract critical manufacturing features from engineering product CAD models. FR technology is used for automating the extraction of data from CAD product models and uses wire-frame geometry extracted from an IGES neutral file format. Existing hint-based feature recognition techniques have been extended to encompass a broader range of manufacturing domains than typical in the literature, by utilizing a combination of algorithms, each successful at a limited range of features. Use of wire-frame models simplifies product geometry and has the potential to support rapid manufacturing shape evaluation at the conceptual design stage. Native CAD files are converted to IGES neutral files to provide geometry data marshalling to remove variations in user modelling practice, and to provide a consistent starting point for FR operations. Wire-frame models are investigated to reduce computer resources compared to surface and solid models, and provide a means to recover intellectual property in terms of manufacturing design intent from legacy and contemporary product models. Geometric ambiguity in regard to what is ?solid? and what is not has plagued wire-frame FR development in the past. A new application of crossing number theory (CNT) has been developed to solve the wire-frame ambiguity problem for a range of test parts. The CNT approach works satisfactorily for products where all faces of the product can be recovered and is tested using a variety of mechanical engineering parts. Platform independent tools like Extensible Mark-up Language are used to capture data from the FR application and provide a means to separate FR and decision support applications. Separate applications are composed of reusable software modules that may be combined as required. Combining rule-based and case-based reasoning provides decision support to the manufacturing application as a means of rejecting unsuitable processes on functional and economic grounds while retaining verifiable decision pathways to satisfy industry regulators.

An approach for developing a computer-based aid to assist in monitoring and assessing nuclear power plant status during situations requiring emergency response has been developed. It is based on the representation of regulatory requirements and plant-specific systems and instrumentation in the form of hierarchical rules. Making use of inferencing techniques from the field of artificial intelligence, the rules are combined with dynamic state data to determine appropriate emergency response actions. In a joint project with Portland General Electric Company, a prototype system, called EM-CLASS, was been created to demonstrate the knowledge-based approach for use at the Trojan Nuclear Power Plant. The knowledge domain selected for implementation addresses the emergency classification process chat is used to communicate the severity of the emergency and the extent of response actions required. EM-CLASS was developed using Personal Consultant Plus (PCPlus), a knowledge-based system development shell from Texas Instruments which runs on IBM-PC compatible computers. The knowledge base in EM-CLASS contains over 200 rules. The regulatory basis, as defined in 10 CFR 50, calls for categorization of emergencies into four emergency action level classes: (1) notification of unusual event, (2) alert, (3) site area emergency, and (4) general emergency. Each class is broadly defined by expected frequency and the potential for release of radioactive materials to the environment. In a functional sense, however, each class must be ultimately defined by a complex combination of in- plant conditions, plant instrumentation and sensors, and radiation monitoring information from stations located both on- and off-site. The complexity of this classification process and the importance of accurate and timely classification in emergency response make this particular application amenable to an automated, knowledge-based approach. EM-CLASS has been tested with a simulation of a 1988 Trojan Nuclear Power Plant emergency exercise and was found to produce accurate classification of the emergency using manual entry of the data into the program. / Graduation date: 1997

The purpose of this research was to study agricultural crop production 'decision support systems' as a means of transferring agricultural technology from research labs and plots to producers, extension specialists, agriculture service agencies, and scientists, on the Western Canadian Prairies. A 'decision support system' is a computer program that analyses problems spanning several knowledge or problem areas producing results that aid the management decision-making process. The primary objective was to develop a computer application program that would fulfill the farm manager's decision support needs and be "open" to future enhancements. This interdisciplinary study has a strong agricultural presence in the application context of the resultant computerized agricultural decision support system, with agronomics being the foundation on which the system was built, and computer science being the toolbox used to build it. Farm Smart 2000 is the resultant decision support system, providing "single-window" access to three different tiers of decision support utilizing the Internet, ' expert systems' and integrated multiple heterogeneous 'reusable agents' in a cooperative problem-solving environment. An ' expert system' is a computer program that solves complicated problems, within a specific knowledge or problem area, that would otherwise require human expertise. Expert systems integrated with each other within a decision support system are called 'agents. Reusable agents' are modular computer programs (e.g. expert systems) which can be used in more than one computer application with little or no modification. Farm Smart 2000 provides support for most management aspects of crop production including variety selection, crop rotations, weed management, disease management, residue management, harvesting, soil conservation, and economics, for the crops of wheat, canola, barley, peas, and flax. Tier-3, the most sophisticated level of Farm Smart 2000, is the focus of this dissertation and utilizes multiple reusable agents, integrating them such that they cooperate together to solve complex interrelated crop production problems. A Global Control Expert achieves the required communication and coordination among the agents resulting in an "open system", enabling Farm Smart 2000 to extend its problem-solving capabilities by integrating additional agents and knowledge, without system re-engineering, thereby remaining an ongoing technology transfer vehicle.

In multi-level decision making (DM) approaches, the final decision is reached by going through a finite number of DM levels. Usually, in each level, a raw decision is produced first and then a suitable decision fusion technique is employed to merge the lower level decisions with the raw decision in the construction of the final decision of the present level. The basic difficulty in these approaches is the determination of how the consecutive levels should interact with each other. In this thesis, two different multi-level DM models have been proposed. The main idea in the first model, &ldquo / hierarchical DM&rdquo / (HDM), is to transfer the decisions of previous hierarchical levels to an upper hierarchy with some reliability values. These decisions are then fused using a suitable decision fusion technique to attain more consistent decisions at an upper level. The second model &ldquo / local DM in multiplelevels&rdquo / (LDM-ML) depends on what may be called as local DM process. Instead of designing an agent to perform globally, designing relatively simple agents which are supposed to work in local regions is the essence of the second idea. Final decision is partially constructed by contribution of a sufficient number of local DM agents. A successful local agent is retained in the agent pool whereas a local agent not successful enough is eliminated and removed from the agent pool. These models have been applied on two case studies associated with fault detection in a four-tank system and prediction of lotto sales.

Traffic accidents which damage the safety of human beings are one of the most important problems due to their material losses and effects to human health. Although continuous improvements are made by the governments / losses of traffic accidents are still a significant issue all over the world. The usual studies realized so far are generally related with the accident prevention models. However, there has not been much research done concerning the situation after the traffic accidents happen. After occurrence of traffic accidents, determination of fault rates for each party involved in the accident is urgently important. The aim of this study is to develop an expert system that uses the knowledge of experts for determination of fault rates in traffic accidents. For this purpose, a detailed literature survey was performed to define the determinants influencing the fault rates of each party. In addition, required data, that is, expert-witness reports were taken from academicians. Classification of these data was done and critical factors affecting fault rates were determined. In light of the defined factors, flowcharts were developed for each type of traffic accident. Moreover questionnaire submitted to experts, was prepared to acquire knowledge of experts. The critical factors affecting fault rates were assessed with a quantitative way in questionnaire. The proposed Traffic Accident Expert System (TAES) is on the basis of the knowledge of experts. Quantification of fault rates can change from one expert to another. An expert system such as the one this thesis will propose will prevent these contradictions. In addition, the expert system quantifies fault rates faster and more consistent as well.

Originally presented as the author's Thesis--Helmut-Schmidt-Universität, Hamburg, 2008. / Includes bibliographical references.

Thesis (doctoral)--Universität, Würzburg, 2000. / Includes bibliographical references (p. [269]-281).

Thesis (M.S.)--West Virginia University, 2004. / Title from document title page. Document formatted into pages; contains vii, 64 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 63-64).

Thesis (M. Sc.(Computer Science))--University of Pretoria, 2003. / Includes bibliographical references.