Daniel R. Rehak

Interfacing expert systems with design databases in integrated CAD systems

Abstract Knowledge-based and expert systems will be important components of future integrated CAD systems. Flexibility and adaptability are of prime importance in incorporating such components into a CAD system. A model of a distributed network DBMS, using knowledge-based programming techniques, for interfacing KBS-to-DBMS is presented. In this model, the description of the data model of each KBS and DBMS component of the CAD system is represented as knowledge description of the data model of each kBS independent of one another. Each component issues and receives data requests in terms of its own data model, and has no knowledge of the syntax, structure and operation of the other components of the system. KADBASE, a prototype of such a flexible interface, under development is described. It demonstrates an approach to developing an integrated, is distributed CAD systems containing a variety of heterogeneous expert systems and design databases.

EVALUATION OF FINITE ELEMENT SYSTEM ARCHITECTURES

Abstract The finite element method is a powerful engineering analysis tool. However, the scope and size of the problems solved are limited by the capabilities and costs of computers currently used. New computer architectures with parallel processing capabilities can exploit the parallelism in the finite element method. Because of the large number of alternative hardware and software configurations and the high costs incurred in the development of such systems, a methodology is needed to compare proposed finite element systems without implementing either hardware or software. This paper presents such a methodology. System architectures are evaluated by simulating the execution of the finite element software on the hardware. Hardware is represented as a set of processors and resources, while the software is modeled as a set of computational tasks organized into an acyclic directed graph. The simulation is performed by scheduling routines. The result of the scheduling routines is a schedule consisting of task-processor assignments and task starting and finishing times. Also, processor and resource utilization levels are generated. These results provide a means of comparing proposed finite element systems.

A knowledge-intensive planner for construction projects

Abstract This paper reviews the capabilities and application of the CONSTRUCTION PLANEX system. This knowledge-based expert system assists in the selection of construction technology, the definition of work tasks, the estimation of required resources and durations, the estimation of costs, and the preparation of a project schedule. The system has been applied in laboratory simulations in three ways: interactively as an intelligent assistant, independently as an automated planner, and as part of an integrated building design environment. CONSTRUCTION PLANEX can plan for the excavation and structural erection of concrete and steel buildings; example applications have ranged from two to twenty stories. Provisions are included in the system for expanding the knowledge base to consider other functional elements or facility types. Descriptions of the system architecture, knowledge representation scheme, control procedures, and applications are included.

Carnegie Mellon Online: Web-mediated education

Distributed education, Web-based instruction, virtual classes-these are all trends in applying technologies to education and learning. One effort at Carnegie Mellon to explore Technology Enhanced Learning (TEL) has been through a system known as Carnegie Mellon Online. Online provides a technical infrastructure to deliver materials to our students both on and off campus. It is designed to deliver educational content and assess student competence in support of courses across the Universitys curriculum. The focus has been large, core, academic courses. We have built a course-neutral, database-driven, student-centered infrastructure that supports both Web-based course delivery and course management. In addition to delivering conventional Web-based hypertext and other media, the system generates customized content (e.g., assessments, feedback) for each individual student and tracks the student through a course while enforcing course specific rules and policies (e.g., topic prerequisites, time and place constraints for exams). The system is designed to be scalable to handle large numbers of courses and students. We describe our work to date, including our pedagogical motivations and our technical approach. We conclude with a discussion of how our approach differs from other uses of the Web in education.

Construction Planex: A Knowledge Intensive Planner for Construction Projects

This paper reviews the capabilities and application of the CONSTRUCTION PLANEX system. This knowledge-based expert system assists in the selection of construction technology, the definition of work tasks, the estimation of required resources and durations, the estimation of costs, and the preparation of a project schedule. The system has been applied in laboratory simulations in three ways: interactively as an intelligent assistant, independently as an automated planner, and as part of an integrated building design environment. CONSTRUCTION PLANEX can plan for the excavation and structural erection of concrete and steel buildings; example applications have ranged from two to twenty stories. Provisions are included for expanding the knowledge base to consider other functional elements or facility types. Descriptions of the system architecture, knowledge representation scheme, control procedures, and applications are included.

Development methodologies for scientific software

Approaches to the development of scientific application software have matured over the past two decades and now constitute identifiable methodologies. Three software development methodologies are described, compared, and contrasted from the viewpoint of development effort, continued maintenance, and subsequent extension. Pertinent software aspects of the POLO‐FINITE system are described to provide an example of the software virtual machine approach to solve database, memory management, and processing module integration problems. Finally, possible extensions of the virtual machine concept are discussed as a means to further advance software development methodologies.

Applications of coarse-grained dataflow in computational mechanics

Dataflow models are free of side effects and have no notion of state or sequencing. Because these representations place a partial, as opposed to a total, ordering on the execution of their component operations, the concurrent aspects of computation are clearly revealed. The correspondence between dataflow graphs and purely functional programs allows computations to be expressed in a high-level functional language and subsequently transformed into a dataflow graph. This paper describes the use of dataflow models as an alternative control strategy for engineering analysis programs and contrasts them with traditional imperative approaches. The characteristics of functional languages are also described, as is their inherent parallelism, which may be realized by compilation into dataflow graphs. The application of functional languages to finite element programming is presented, which allows the alternating assembly and solution of system equations found in frontal solvers. Issues such as the incremental update of arrays and the simulation of state are also addressed.

An Expert System for Construction Planning

Among other concerns, construction planning involves the choice of construction technology, the definition of work tasks, the estimation of required resources and durations, the estimation of costs, and the preparation of a project schedule. A prototype knowledge intensive expert system to accomplish these tasks, CONSTRUCTION PLANEX, is described in this paper. This system generates project activity networks, cost estimates and schedules, including the definition of activities, specification of precedences, selection of appropriate technologies and estimation of durations and costs. The CONSTRUCTION PLANEX system could be useful as an automated assistant in routine planning, as a laboratory for the analysis and evaluation of planning strategies, and as a component for more extensive construction assistance systems involving design, site layout or project control. The current application for CONSTRUCTION PLANEX is to plan modular high rise buildings, including excavation, foundation and structural construction.

Expert systems and CAD databases

Database management systems and expert systems will be important components of integrated computer-aided design systems. A powerful, adaptable interface between these components Is necessary to build a comprehensive CAD system. The paper discusses the basic Issues involved in interfacing expert systems with database management systems and describes the architecture of a prototype system, KADBASE, that is currently under development. KADBASE is a flexible, knowledge-based interface In which multiple expert systems and multiple databases can communicate as Independent, self-descriptive components within an integrated, distributed engineering CAD system. The paper presents a detailed example to demonstrate the use of KADBASE in a typical engineering design application.

Geometry and Domain Modelling for Construction Robots

Construction robots operate in an environment very different from that of manufacturing robots. This environment is less structured, more complex and more dynamic than is the norm in manufacturing. In addition, construction robots are inherently mobile, as they are engaged in building or maintaining an immobile structure which is large compared to their dimensions. Another complicating factor is the uniqueness of actions that must he taken by a construction robot: the number of special conditions that may exist in buildings is large. All of diese differences provide arguments for two related capabilities that arc required of construction robots: the ability to reason about and to model their environment. In diis paper we present current work at Carnegie-Mellon University which addresses the problems of geometrie reasoning and domain modelling in the specific context of knowledge based expert systems.