Knowledge Acquisition for Bridge Design Expert Systems

Abstract: Knowledge acquisition for developing design knowledge bases is difficult and time consuming. In this paper, a new approach for acquiring design knowledge is presented. The object-oriented approach is applied to model the design objects and the design process. It is discussed that in the case of routine bridge design, synthesis of the bridge structure can be achieved through recursive top-down refinement using selection among predefined types, and decomposition of the structure into predefined components. Having this design model, a methodology for using analogy to augment the knowledge base is proposed. Analogy can be used to retrieve and modib an available design object in order to create a similar object interactively. A prototype system that implements this methodology is presented.     

浮法玻璃退火窑故障诊断专家系统设计

开发了一种浮法玻璃退火窑故障诊断专家系统,能有效解决目前浮法玻璃退火窑故障诊断随意性和滞后性的问题.该系统采用Visual C 语言编写,增强了用户界面的友好性,便于用户在较短时间内迅速掌握此系统的运用.编写过程中采用了实例推理机制,提高了结论的可靠性及运行速度.     

Extending Prolog to Provide Better Support for Design Code Expert Systems

The CSIR0³ Division of Building Research (DBR) has built a number of text-based and graphics-based design code expert systems in PROLOG. Each of these systems involves several thousand lines of PROLOG and stretches to the limit and the capacity of the IBM ATs on which they run. We have found PROLOG to be a very powerful language for expressing the complex rules and relationships found in building design codes and related applications. Nevertheless, it has a number of weaknesses, particularly in expressing procedural knowledge and arithmetic relationships, and in implementing complex tables and user interfaces. We have designed and implemented extensions to PROLOG to overcome these difficulties. These include new built-in functions to perform functions that are difficult to code in PROLOG, an easy way to add further built-in functions, a meta-interpreter to keep track of intermediate results, sets of routines to support user input and explanation, and a definite clause grammar to implement a more syntactically pleasant language for writing rules. These extensions are being used to develop a larger and much more complete version of the WINDLOADER expert system, and have greatly facilitated this task.     

A Local Database Manager for Engineering Analysis Systems

Abstract: A local database manager (LDBM) is presented. It is a software development tool consisting of a set of easy-to-use in-core data management subroutines written in FORTRAN 77. One of the main motivations for the development of this LDBM was to enhance the eflcient use of core memory which seems to be obvious with respect to a PC programming environment. However, even on virtual machines, high performance achieved by minimizing page faults is strongly based on a sophisticated database management system (DBMS). In order to accomplish this objective, the LDBM allows the use of different data types according to the requirements of the actual situation and to the existing programming environment. Integer, short integer, real, double precision, complex, double complex and character data can be handled. Moreover, a two-level hierarchical data structure is used which groups related data in a different and more economic way rather than by means of increasing the number of levels in the DIMENSION statement. Of course, old, no longer used data may be deleted, leaving room for the allocation of new data. Using a technique of 'flexible garbage collection', i.e. the ensuing gap after the deletion of a data block is not automatically closed, the time consumed by the LDBM during a deletion operation is actually not affected by the length of data blocks.     

Structuring Knowledge and Data Bases in Expert Systems for Integrated Structural Design

The integrated structural design includes preliminary design, structural analysis, and the final detailed design. A knowledge-based expert system for integrated structural design requires both symbolic processing of heuristics and experiential knowledge and numerical processing of algorithmic computations. In addition, such a coupled expert system must use existing data bases as well as a large variety of data created during a consultation with the system. Thus, considering their complex interaction, effective integration of symbolic processing, numerical processing, and data base management becomes a major concern in the development of expert systems for integrated structural design. This paper addresses this integration problem by presenting a prototype system, called FRAMEX, for design of a class of steel structures—regular multistory steel buildings. The basis of steel design is the recently developed Load and Resistance Factor Design Specification.     

A Knowledge-based Expert System for Design of Roof Trusses

Aknowledge-based expert system has been developed for design of roof trusses, called RTEXPERT. RTEXPERT can advise the user on the appropriate type of the roof truss, selection of the layout of the truss (such as the pitch of the truss and number of panels), and the loading. The basis of design is the American Institute of Steel Construction (AISC) specification. The truss is designed for dead, live, snow, and wind loads in accordance with the American National Standard Institute (ANSI) specification. A novel part of RTEXPERT is the automatic computation of nodal forces due to various loads. The user needs to indicate only the types of the materials used as roof covering and the location of the structure in the United States. RTEXPERT automatically generates all the nodal forces. The knowledge base and explanation facility of RTEXPERT have been developed using INSIGHT 2 + expert system shell. The mathematical computations, graphic algorithms, and data file manipulation routines have been developed in Turbo Pascal. RTEXPERT has a comprehensive graphic interface for displaying the truss configuration, cross sections, loading, and deformed shape. Information about individual members is presented through multi-window graphicstext displays. RTEXPERT can be used as an "intelligent assistant" for design of roof trusses.     

液压系统故障诊断专家系统

本针对液压系统故障诊断问题，根据系统和故障的分级特性，建立了相应于“深浅知识库”的双层故障诊断模型，给出了针对这种模型的知识处理方法，讨论了推理中为解决不确定性而使用的模糊综合评判方法。     

Uncertainty Measures in Expert Systems

Abstract: In classic logic, premises, conclusions, and rules are treated deterministically, i.e., they are considered as either true or false. However, when dealing with reality, one finds that all these elements must be considered as uncertain. Thus classic logic must be extended to cover real situations. One possible extension is given by uncertainty measures together with aggregation formulas that combine the uncertainty of premises with that of the rules to obtain the uncertainty of conclusions. This paper describes different uncertainty measures, giving the physical meaning of the implied axioms and their limitations, illustrated by some examples. Finally, a classification of some of the well-known uncertainty measures, such as belief and plausibility functions, probabilities, necessities, and possibilities, is given.     

Uncertainty Methods in Expert Systems

The paper introduces the main measures of uncertainty and discusses some of the problems associated with conventional uncertainty propagation methods, such as those based on independent probabilities and the methods used with certainty factors, belief, and possibility functions. Some examples show the importance of the associated errors. Alternative methods, such as log-linear regression and causal networks or influence diagram models, are discussed. Finally, their structural and parametric learning possibilities are analyzed.     

Linear Programming and Expert Systems

Abstract: This paper describes an expert system shell based on linear programming. The knowledge base consists of the parametric equations of the feasible set, which is obtained from upper and lower bounds of probabilities of given sets. The inference engine consists of an algorithm that gives the parametric equations of the feasible set and works in an incremental manner; any previously calculated feasible set is modified when new information becomes available, using all previous calculations. This allows for rapid updating of the knowledge base and permits a very quick solution of the optimization problems (finding of lower and upper bounds of the probability of any set) to be obtained. Finally, three civil engineering examples are included.     

Log-Linear Models in Expert Systems

Abstract: The paper describes how log-linear models can be used to deal with uncertainty in expert systems, avoiding the common problems of many probability-based expert systems. After a general introduction to log-linear models, including hierarchical models, maximum likelihood estimation for poissonian and multinomial sampling is described and parametric and structural learning methods are illustrated by simple examples. Finally, a traffic engineering example is given.     

An Expert System for Tunnel Design

In this article, we describe the development of an expert system (ES) for tunnel design and preent some applications of the system. The system has been developed by our research group consisting of 21 people closely involved with the tunnel design under the leadership of Nagoya University. The system consists of four subexpert systems (1) ES for standard tunnel design methods, (2) ES for framed structure method, (3) ES for theoretical design methods, and (4) ES for numerical analysis design method; a common part to control the overall system; and two data base systems (tunnel data base system and rock mass data base system).     

Expert Systems for the Interpretation of River Quality Data

Sample data on river quality are used for a variety of management purposes. The paper considers the role of expert systems in interpreting such data. An example is given of a prototype rule-based system designed to aid in assessing compliance with European Community (EC) Directives. It is suggested that, although this type of formal characterization is a necessary part of management, it provides poor information for decision making. The development of an expert system to provide a more accurate and informative interpretation of episodic pollution events is then described. The paper concludes by discussing the practical application of these approaches.     

浅谈城市设计中的界面设计——以南宁市邕江北岸(北大桥—心圩江)城市设计为例

通过对南宁市邕江北岸（北大桥—心圩江）城市设计中城市界面设计控制与引导以及城市界面管理方面的思考,寻求可持续发展的、务实的解决方案,以期指导城市的开发建设,为今后的城市界面设计以及规划管理提供有益的探索.     

Combining Expert Systems and Cad Techniques

Drawings provide a medium for communicating design decisions. The dtask of preparing and changing design drawings has been simplified by various packages available for computer aided drafting (CAD). Recently, expert-systems techniques are being used to aid the design process by monitoring and refining design decisions. Combining expert systems and CAD techniques raises issues concerning representation of the design. An expert system for evaluation of the design. An expert system for evaluation of preliminary structural designs has been implemented using an expert system shell and a CAD package. This and other similar implementations illustrate different approaces to the integration problems     

Causal Network Models in Expert Systems

Abstract: The paper describes causal network models, giving a detailed discussion of their basic assumptions and their associated algorithms for marginalization and updating. Conditional probability tables are shown to be the basis for the aggregation formulas of if-then rules allowing for deriving the uncertainty of conclusions from that of premises. Examples of a pressure tank and a power system are used for illustrating the concepts and all steps of explained algorithms.     

Expert Systems and Documentary Dataases Integration

Abstract: The CSTB has been involved for many years in the process of a better availability and better use of building regulations. The FARTEC project has provided a significant breakthrough in that direction, thus offering a digitized corpus and its software environment, facilitating the retrieval and the handling of technical and legal texts. A related research efort, referred to as the ACSIS project, aims at devising advanced software environments where building standards represent the basis of intelligent softwares, moreover including a wide range of functions such as technical control, automated selection of manufactured products, assisted designs, etc. The ACSIS system offers various access modes to the corpus (indexation, graphical access, local dictionaries) available for different data types (text, images, arrays, formula, etc.), and associates this primary information to intelligent softwares (waterproofing expert adviser, anti-seismic design controller, accessibility project checker for disabled people, etc). Sophisticated dynamic indexation processes are offered to enable the user an eficient selection of the documents given a set of nomenclatures evolving according to the current user context. Moreover, each one of the high added value applications connected to the corpus is able to make an appropriate use of the building standards stored in the SGML primary corpus. Given a set of Document Type Definitions (DTD), the object-oriented document model and the hyperobject networks for each document of the SGML database are created: this stage is known as the compiling of the corpus. Then, given a set of formatting directives (e.g. indentation, justification, ere.>, and of presentation requirements (e.g. fonts, etc.), labelling each DTD structural entity (e.g. article, title, paragraph, etc.), the visual display of the documents is produced. Finally, the documentary database is linked to the intelligent softwares, thus offering a crossvalorization of the expert knowledge, illustrating, on one hand, the foundations of the expert rules and, on the other hand, the appropriate operational use of the regulation.     

An Expert System for Auditing Quality Management Systems in Construction

Abstract: This article introduces a system called Construction Quality Management Audit (CQMA) Expert that assesses the performance of a quality management system (QMS) implemented in a construction firm. CQMA Expert is programmed by using MATLAB's GUI components and its Fuzzy Logic Toolbox. CQMA Expert's rule base is constructed using information obtained from auditors of QMSs. CQMA Expert imports the quality requirements relative to the many quality management processes specified in ISO 9000, processes audit inputs, and generates consistent decisions relative to conformance to standards. It provides an interactive user interface for recording evidence collected during the audit and clearly states the reasons for the conclusions. It contributes to continuous quality improvement because (1) it enhances the maintenance of a QMS by quantifying its performance, (2) it assists with and facilitates the implementation of the duties of auditors in charge of assessing the performance of a QMS, (3) it simplifies the burdensome process involved in keeping track of the audit results of the many quality management processes investigated, and (4) it reduces the impact of the variability caused by the use of different auditors assessing different quality management processes. Case studies based on Section 4.11 of the ISO 9000 standard entitled “Control of inspection, measuring and test equipment” are used to illustrate the system and verify its usability and validity.