Showing papers in "Journal of Computing in Civil Engineering in 1996"

Fuzzy Controlled Genetic Algorithm Search for Shape Optimization

Abstract: Shape design of a structure plays a significant part in deciding its least weight. In this paper, a fuzzy controlled genetic-based search technique for structural shape optimization is investigated. An automated optimal procedure based on the proposed approach is developed and used in the least-weight design of truss structures, which include their geometry as a design variable to be optimized. To increase the performance of the genetic-based approach for shape optimization problems, the design constraints related to member stress, joint displacement, and member buckling are described by using fuzzy set theory. A fuzzy rule-based system representing expert knowledge and experience is incorporated in the approach to control its optimal search process. Four examples for shape designs are presented to demonstrate the effectiveness and efficiency of the proposed hybrid approach in comparison with the use of pure genetic algorithms and other numerical methods. The examples show that the approach is flexible enough to deal with rigidly jointed structures.

Models of Construction Process Information

Abstract: Computer-integrated construction (CIC) and concurrent engineering for architecture, engineering, and construction (AEC) require data standards or common information models through which computer systems can exchange project information High-level conceptual core models are required as unifying references for the more detailed, application-specific models used for the actual information exchange A variety of core models have been developed in the area of AEC process information This paper introduces several such models from a variety of projects It discusses general modeling mechanisms and looks at several issues relating specifically to process modeling for AEC, comparing the approaches of the various core models described, and providing some recommendations The overall objective is the eventual emergence of generally accepted standards in this area

Reliability tester for water-distribution networks

Abstract: A computer model for assessing the reliability of water-distribution networks is presented. Both mechanical failure caused by pipe breakages and hydraulic failure caused by insufficient pipe capaci...

Object-Oriented Model for Integrating Construction Product and Process Information

Abstract: Product and process models provide the necessary information framework for implementing computer systems for the architect/engineering/construction (A/E/C) industry. Although the focus of these models is slightly different, both are needed to provide a foundation for managing project information during the design and construction phases. Design information— “product” information based on building components—needs to be integrated with construction management tasks, the “process” information necessary to build the components. It is therefore important to provide an integrated information model to bridge the gap between product and process information for a construction project. An integrated information model not only encourages those involved in construction to use and add to design information, but also provides richer information representation, better efficiency and data consistency, and the flexibility to support life-cycle information management. The research presented in this paper was performed under the auspices of the collaborative engineering research program at the U.S. Army Corps of Engineers Construction Engineering Research Laboratories (USACERL), which is attempting to redefine existing design processes to make them more collaborative and to develop enabling technologies to support the new process. An important part of this research is the development of an integrated information model that allows agents to communicate/collaborate over the life cycle of the project. This paper presents an object-oriented model that integrates product and process information to support collaboration among design and construction agents, and two prototype construction agents for construction planning and monitoring project progress. The development of these two agents demonstrates the value of using integrated product and process models for managing facility information in the A/E/C industry.

Examples and characteristics of shared project models

Abstract: Integration of data and knowledge among project participants and their computer applications promises to improve the quality and efficiency of the project-delivery process in the architecture, engineering, and construction (AlFJC) industry. Integration, we argue, can be achieved through shared project models. In the present paper, we give two examples of such models by summarizing two research projects that developed and used shared project models. In the first example, a project model supports constructability evaluation of a building structure. The second example shows how a shared project model supports construction-management tasks. From these examples, we derive and discuss essential, general characteristics of shared project models. These characteristics include a model's layers of abstraction, content, extensibility, and support of integration mecha­ nisms. These mechanisms range from file sharing, to shared databases, and application-programming interfaces. They support information sharing via custom translation, translation to a shared model, or adoption of a common model at the application, system, or industry level. This paper ends with brief discussions of standardization efforts and implementation technologies.

Dynamic Programming Approach to Scheduling of Nonserial Linear Project

Abstract: This paper presents a nonserial dynamic programming formulation for the scheduling of linear projects with nonsequential activities. One of the major advantages of the approach presented is its ability to handle both serial and nonserial linear projects with activities performed with variable crew formations. The presented formulation determines the optimum crew size for production activities that lead to the minimum project total cost. An example project is provided in order to illustrate the computational steps, validate the calculation algorithm, and show the capabilities of the proposed method.

Predicting dynamic response of adsorption columns with neural nets

Abstract: Artificial neural networks have been applied to a limited number of environmental engineering problems. This paper investigates the feasibility of utilizing the concept of neural nets in developing networks for predicting the breakthrough curves of fixed-bed adsorbers. Close agreement is observed between the breakthrough curves predicted by the developed neural network and those obtained from the mathematically based adsorption model (HSDM). The advantages of using neural networks in modeling environmental processes over the commonly used traditional methods are addressed. Also, further improvements and generalization of the developed predictive fixed-bed adsorption neural network are discussed.

Network Expert Geographic Information System for Landfill Siting

Abstract: A prototypical network expert geographic information system (GIS) is developed to facilitate municipal solid waste landfill siting. A forward chaining knowledge base consisting of related siting rules extracted from various literature is used to establish an expert system. Siting analysis is performed by a GIS and evaluated by rules triggered from the expert system. The expert system and GIS are integrated into an expert GIS to combine the advantages of both systems. Also, a multimedia network interface is designed for local or remote access to the system from anywhere on the Internet. With the interface, a general user does not require previous knowledge of the expert system and GIS to use the system. This networking capability not only significantly reduces the cost of distributing tools to each user, but also avoids the difficulty that each user may encounter in installing and managing tools on the computer. Finally, a case study is provided to demonstrate the effectiveness of the developed system in facilitating landfill-siting analysis. The system can be directly accessed via Internet from the home page http://ev004.ev.nctu.edu.tw/ENGLISH/wsite/index.html.

Geophysical Log Interpretation Using Neural Network

Abstract: Timely and effective interpretation of bore hole geophysical and formation well logs is vital in developing basic geological and hydrological data for ground water modeling. Information on local geological conditions may be estimated from many types of geophysical and formation logs; however, interpretations of these data can be subjective and time-consuming. A trained neural network can be used effectively and efficiently to complement manual log interpretation. In this paper, a neural network is developed to analyze geophysical well logs and to provide information on the subsurface strata classifications. An analysis is given on the neural network development process and data requirements. An overview is presented on the neural network optimization techniques, limitations, and the strength of the approach in well-log interpretation.

Building process models for design management

Abstract: Traditionally, computers have been used as tools to enhance productivity on an individual basis. Integrated process support environments (IPSEs) change this work paradigm. Within an IPSE, the wider business processes of communication, cooperation, collaboration, information exchange, triggering of activities, etc. are enacted within a networked computer environment. Participants become inhabitants in a process that is partly managed by the computer environment. In the new paradigm technology is a driver of change rather than an enabler. Consequently, the introduction of process technologies presents organizations with an opportunity to reengineer the way they work. However, the development of methodologies to support process technologies is still immature and underdeveloped. The findings of a study aimed at developing process understanding for design management within two large civil engineering organizations are reported here. The study used the learning cycles of soft systems methodology (SSM) and grounded theory to guide the process. Embedded into these methodologies were the process-centered approaches of walk-through scripts, role developments, and role activity diagrams (RADs). Examples of how these approaches were developed in practice are shown.

Decision-support system for infrastructure preservation

Abstract: This study presents the development of a decision-support system for the preservation of civil infrastructure. The system aims at providing assistance for decisions concerned with the three main ta...

System for bridge management in a rural environment

Abstract: A system is presented for the cityor county-level management of bridges located in Wyoming. Although the research work was directly focused on the state of Wyoming, the consideration involved in the system development, the procedures involved, and the final software design are generally applicable to other states that have similar demographics and bridge inventories. Selected bridge management systems (BMS) are reviewed as appropriate, and characteristics of each, along with their advantages and disadvantages, are elaborated. Several of the favorable characteristics of these systems were modified and incorporated into the present system, called WYO_BMS. The objective of this paper is to provide the general background for the development of the WYO_BMS. Details specific to the Wyoming implementation are not provided here, but are available in the final report.

Computerized tool for hierarchical simulation modeling

Abstract: Advancements in computing have enabled the development of effective tools for the modeling and analysis of construction operations. In particular, recent developments in object-oriented and event-driven programming were used by the writers in the development of a hierarchical, modular, and graphical environment for simulation-based planning of construction projects. This system, termed hierarchical simulation modeling (HSM) method, is specifically geared toward modeling the uncertain nature of conditions under which projects are implemented and the dynamic utilization of resources in construction. The modeling concepts, internal structure, implementation, and working of this environment are described in this paper. “Model definition,”“translation,”“simulation,” and “report generation” are the four internal modules of this system. One of the key concepts highlighted in this paper is the use of an object-oriented paradigm for developing modeling objects that have “graphics” and “simulation” properties that ...

Multigrid methods in gis grid-cell-based modeling environment

Abstract: Multigrid methods use a series of successively coarser grids to accelerate the solution of equations arising from the numerical solution of partial differential equations A grid-cell-based geographic information system (GIS) with a map algebra language is capable of displaying and manipulating spatial data and attributes and handling the data arrays and results arising in ground-water simulation problems To examine the efficacy of GIS to solve ground-water flow problems, we have performed ground-water modeling directly within a GIS without going outside the system for model solution We have applied the multigrid method to solve steady ground-water flow problems in heterogeneous aquifers Execution times for the GIS-multigrid approach using simple iterative techniques greatly exceed those of traditional ground-water simulation models due to the creation of temporary grids for intermediate calculations during the simulation Efficient handling of boundary conditions is another difficult problem in GIS grid-cell-based modeling

Average and peak traffic volumes: neural nets, regression, factor approaches

Abstract: Improving the quality of predictions from sample data is an important aspect of a data collection and analysis program Regional and environmental agencies use different statistical modeling techniques for prediction The advent of neural network technology has introduceed a new set of prediction models It is important to make sure that the modeling technique used provides the best possible accuracy This study compared the existing approaches to the estimations of average and peak hourly traffic volumes with the multiple regression analysis and the neural network approach All the approaches were compared using different classification schemes as well as different durations of traffic counts The multiple regression analysis and the neural network approaches consistently performed better than the conventional approach Apart from suggesting good modeling tools for estimating average and peak hour traffic volumes, the results also provide useful insight into the durations of short-term traffic counts and the classification schemes for highway sites

Autonomous Mapping System for an Interior Finishing Robot

Abstract: The purpose of an autonomous-control system for interior finishing robots is to map an unknown building interior, plan the performance of a desired task, and monitor its execution by the robot. Two types of mapping systems are examined: one is based on a laser range finder, the other on a combination of an ultrasonic sensor for mapping of walls with an infrared sensor for mapping of openings. Mapping procedures with both systems of single rooms of different configurations, and of a whole building floor are explained. The characteristics of the sensors involved in both systems were tested in a real building interior environment. The results of these physical experiments were fed into computer programs developed to simulate the entire mapping procedure. The accuracy of the mapping and its economic feasibility are also examined.

Multidisciplinary Product Modeling of Buildings

Abstract: A building entity is decomposed into systems (containing subsystems and elements), which correspond to the various disciplines associated with it. A set of exhaustive, nonintersecting primary elements are identified for the architectural and structural systems. These are space, solid, opening, layer, and finish. The classification of solids into principals and joints provides a partial solution to the problem of different disciplines having different views of building entities. Joints can also be used to define nodes, act as load transfer junctions and store information (e.g. hogging moment information for beams). Structural engineering perspectives and processes require that principals and joints be aggregated into assemblies. Juxtaposition between the preceding primary elements is handled by the definition of three distinct interface elements, namely face, surface, and boundary. The building product model described before is implemented in a limited fashion in a PC environment on an object-oriented shell. The example provided shows how the model can be used for processes such as load evaluation, structural analysis, design checking, and quantity takeoff.

Object orientation in hydraulic modeling architectures

Abstract: A “separable” approach to the object-oriented design of modeling systems is presented. In it, the local behavior of an individual object is separated from the global behavior of the system, and the functional decomposition of the system is separated from its synthesis as a set of interacting objects—entities that encapsulate data and behavior. In a pipe network, elements such as pipes, junctions, pumps, and valves can be represented in the form of objects (so-called agents) encapsulating specific attributes and the behavior of each element. The behavior of the network as a whole is encapsulated in an object that controls agents (so-called actor). This approach is demonstrated in the design and implementation of an interactive water distribution modeling and design system, the hydraulic modeling system (HIS); an example of its application is given. The use of object orientation in constructing computer-based models is viewed as very promising in the field of hydraulic and environmental modeling, and especi...

Finite-Element Graphic Objects in C++

Abstract: This technical note describes the design and application of a group of graphical finite-element object classes that were developed using the object-oriented approach and were implemented in C++ These object classes were designed to encapsulate the characteristics of finite-element objects and corresponding graphic drawing functionalities Thus, graphic functionalities can be incorporated easily into a finite-element analysis program by the selective use of these object classes Extension of these object classes can lead to new finite elements with different functional and graphical characteristics A windows-based finite-element analysis program with an interactive graphic user interface was developed using these object classes as basic building blocks The \Iinheritance\N feature of the object-oriented programming (OOP), which allows some codes of existing classes to be reused, was utilized to simplify the programming process To demonstrate the \Ireusability\N and \Iextendibility\N characteristics of the object classes, a different finite-element analysis program was coded by using the same object classes

Object-oriented analysis of South Florida hydrologic systems

Abstract: Object-oriented analysis is a methodology used to study real-world systems by identifying system components and behavior. An object-oriented analysis of south Florida hydrologic systems was conducted to determine how well this methodology organizes system information. The principal products of this analysis were three conceptual models characterizing the overall structure, behavior, and functionality of south Florida hydrologic systems. The object model identifies system components, or classes, their attributes, and their relationships with one another. The dynamic model presents class behavior, and the functional model shows how class attribute values are transformed. These models represent a variety of south Florida hydrologic systems and can be used to develop design specifications for a library of modular software objects, in which each object corresponds to a class in the object model.

Knowledge acquisition and engineering for steel bridge fabrication

Abstract: The Kansas Department of Transportation (KDOT) developed a knowledge-based system for resolving fabrication errors in steel highway bridges. The system--Bridge Fabrication Error Solution Expert System (BFX)--focuses on errors occurring with relative frequency, but not resolved by standard actions prescribed by KDOT specifications or guidelines. The authors of this paper collected example cases from fabrication shops, state inspectors' field notes, and bridge-project documents. Panel meetings and interviews were conducted with multiple-domain experts. These cases became the foundation for constructing the knowledge representation space and testing of the system. Although an ID3-based learning tool for knowledge acquisition was attempted, it was unsuccessful, leaving knowledge acquisition based primarily on explicit domain extraction. Findings revealed that BFX offered proper repair solutions two-thirds of the time and correctly reported that no solution could be properly generated using the existing knowledge base for the remaining one-third of test cases.

Abstracting Lessons Learned from Design Reviews

Abstract: The application of lessons learned during construction to future designs is a goal of many construction organizations. This paper presents the current status of design review systems within the U.S. Army Corps of Engineers, and describes a new computer program, the Lessons-Learned Generator, that abstracts frequently used design review comments. Frequently used comments in the Reviewer's Assistant system are evaluated based on usefulness, generality, and content stability. Well-formed comments are then abstracted and included in a new Reviewer's Assistant project. The abstraction process keeps the growth of the database at a reasonable level and allows users to identify and distribute sets of repetitive comments. Difficulties associated with the automated abstraction of lessons learned are discussed. Since the Lessons-Learned Generator identifies lessons learned as part of a reviewer's daily business practice, then makes the lessons-learned immediately accessible within the Reviewer's Assistant system, this approach may be more effective than paper checklists or stand-alone automation approaches for applying lessons learned.

Modeling and Debugging Engineering Decision Procedures with Machine Learning

Abstract: This paper reports on the use of machine learning programs for modeling existing engineering decision procedures. In this acitivity, different models of a decision procedure are constructed by using different machine learning programs as well as by varying their operational parameters and input. These models serve to focus on different aspects of the decision procedure thus improving its understandability, which, in turn, can assist in its evaluation and subsequent debugging. This important modeling role of machine learning programs is exemplified by modeling an existing decision procedure that is used by engineers when they need guidance in selecting among available techniques for modeling ground-water flow in a process of environmental decision making. This decision procedure was corrected and improved in the course of this work. The example demonstrates the practical utility of the modeling role of machine learning for engineering applications.

Frameview: Object-Oriented Visualization System for Frame Analysis

Abstract: Rapidly developing desktop computing capabilities, which include high resolution graphics and interactive graphical user interfaces, are leading to a new generation of engineering software. One of the challenges in engineering software development is the effective use of computer graphics for visualization of data. Object-oriented methodologies hold the greatest promise to support reuse and rapid prototyping for large-scale software development. This paper outlines some of the essential characteristics of \HFrameView\N, an object-oriented software system for visualization of responses from any type of frame analysis model. FrameView provides a graphical user interface that is based on the X-Window system and Motif. The graphics library used is PEX, which is a three-dimensional (3D) extension to the X-Window system. The programming language used is C++. Important attributes of PEX, and tedious programming details required when using PEX, have been encapsulated in a number of classes. General classes are developed to abstract the tools needed for viewing frame-element responses and two-dimensional and 3D graphics modeling. Viewing of results from 2D frame elements based on an assumed cubic transverse displacement field for drawing the deflected shape, and an assumed linear function for any other general response quantities along the element length, have been incorporated in FrameView. An example of how FrameView has been extended to handle other graphical representations of frame-element responses is discussed.

Multimedia Development Software: Object-Oriented Interface-Based Simulation

Abstract: This paper provides guidelines for choosing software to develop multimedia educational and training applications. It classifies five levels of application interactivity and considers them as the basis for software categorization. The paper illustrates fully interactive application development through an object-oriented interfaced-based simulation exercise in the area of construction equipment. It finally closes with further considerations for development software based on organizational needs and market analysis.

Synthesized Images for Pavement Management System Design

Abstract: In visual pattern recognition, researchers often rely on standard test images obtained under fixed conditions because it is usually difficult to obtain images with all the desired characteristics. This practice can compromise an investigation because existing images may need major changes in optical parameters or the original scene may be unavailable. For most engineering applications, algorithms can only work well under certain conditions. New algorithms might require specifications not fully met by the existing data (patterns). Synthesized patterns, in which a computer fills these requirements, may be the answer to the problem, due to its advantages of low cost, easy specification, independence from environmental conditions, and realistic visualization of abstract ideas that stimulates creativity. Ray tracing--a computer graphics technique with optical correctness requiring a few numerical parameters--can be employed to devise a scene. Results are presented in the form of light intensity maps, convenient for pattern recognition algorithms. Once the algorithms under study are confirmed with the synthesized images, conditions can be specified for acquiring real-world images, thus reducing costly trial-and-error methods. This paper proposes a novel approach to generate cracks and potholes using "fractals" with "texture mapping." The results illustrate the practicality of the proposed approach in the design and implementation of pavement management systems.

Development of an Expert System for Daily Drought Monitoring

Abstract: A daily drought-monitoring system was developed using the Clips Expert System software package. The monitoring system consists of four components: (1) the knowledge base, which contains rules in th...

Linked lists for transport simulations using Lagrangian parcels

Abstract: The linked-list method is explained using transport simulations involving Lagrangian discrete parcels. Parcel insertion and removal as well as storing of variables in a multispecies problem are explained through examples. The advantages of the linked-list method over the sequential-list method are discussed. Both single-species and multispecies applications are addressed. It is shown that for maintaining parcel lists, insertions, and removals, use of the linked-list approach optimizes the use of memory and is significantly more efficient than using sequential lists.

Sequence control for integrated structural design models

Abstract: An initial step in developing a computer-integrated system for structural design is to create a formal description of the information and activities involved in design. The formal description provides a clear understanding of and a theoretical basis for the computer-integrated design system before the system is implemented. A design model based on entities and relationships can be used to formally describe design information and activities. Product entities describe the information produced during the design process, and process entities describe associated activities. The relationships among the entities of a design model include organizational, interaction, and sequence relationships. Previous research has developed formal concepts and notation for representing organizational relationships such as aggregation/decomposition and generalization/specialization. This paper focuses on the representation of sequence relationships. The sequence relationships of a design model describe the possible sequences in which the process entities (representing design activities) can be initiated during the design process. This paper outlines sequence relationships in a model for structural design and presents a formal notation for representing sequence relationships based on context-free grammars.