John C. Kunz

PUFF: an expert system for interpretation of pulmonary function data

The application of artificial intelligence techniques to real-world problems has produced promising research results, but seldom has a system become a useful tool in its domain of expertise. Notable exceptions are the DENDRAL (1) and MOLGEN (2) systems. This paper describes PUFF, a program that interprets lung function test data and has become a working tool in the pulmonary physiology lab of a large hospital. Elements of the problem that paved the way for its success are examined, as are significant limitations of the solution that warrant further study.

A Trajectory for Validating Computational Emulation Models of Organizations

Validation of complex simulation models is a challenging problem in computational organization theory research. In this paper, we describe a validation strategy suitable for emulation simulation systems, and show how a comprehensive validation consists of a sequence of steps that evaluate different aspects of a computational organizational simulation model. We demonstrate how this strategy can be applied to the evaluation of the Virtual Team Alliance (VTA), an emulation simulation system designed to guide managers charged with organizational change. VTA required a &201C;trajectory&201D; of successive validation experiments, before managers where willing to follow the recommendations of VTA. Ultimately, we believe this validation approach can be applied to a wide range of different simulation systems, and will make identification of the strengths and weaknesses of organizational simulations easier.

A generic feature-driven activity-based cost estimation process

Understanding how the building design influences construction costs is a challenging task for estimators. Estimators must recognize the design conditions that affect construction costs and customize the cost estimate accordingly. Estimators have different preferences for how and when to adjust a project’s activities, resources, and resource productivity rates that form the basis of a cost estimate. Current tools and methodologies lack ways to help estimators customize construction cost information according to their preferences and the particular features in a given design. This paper describes the process we formalized to customize a project’s activities, resources, and resource productivity rates based on a project-independent representation of estimators’ rationale and a project-specific feature-based product model. The formal process creates an integrated model that explicitly relates features, activities, resources, costs and the estimator’s rationale. Our tests show that this process enables a software prototype to generate and maintain cost estimates quickly, consistently, and accurately for feature-based product models. q 2003 Elsevier Ltd. All rights reserved.

A feature ontology to support construction cost estimating

Construction cost estimators are confronted with the challenging task of estimating the cost of constructing one of a kind facilities. They must first recognize the design conditions of the facility design that are important (i.e., incur a cost) and then determine how the design conditions affect the cost of construction. Current product models of facility designs explicitly represent components, attributes of components, and relationships between components. These designer-focused product models do not represent many of the cost-driving features of building product models, such as penetrations and component similarity. Previous research efforts identify many of the different features that affect construction costs, but they do not provide a formal and general way for practitioners to represent the features they care about according to their preferences. This paper presents the formal ontology we developed to represent construction knowledge about the cost-driving features of building product models. The ontology formalizes three classes of features, defines the attributes and functions of each feature type, and represents the relationships between the features explicitly. The descriptive semantics of the model allow estimators to represent their varied preferences for naming features, specifying features that result from component intersections and the similarity of components, and grouping features that affect a specific construction domain. A software prototype that implements the ontology enables estimators to transform designer-focused product models into estimator-focused, feature-based product models. Our tests show that estimators are able to generate and maintain cost estimates more accurately, consistently, and expeditiously with feature-based product models than with industry standard product models.

Designing and Evaluating Visualization Techniques for Construction Planning

Construction project teams view project information with traditional paper-based methods that have remained largely unchanged with the advent of computers and electronic project information. Observations of project teams show that these methods fail to support critical group decision-making tasks because they do not communicate relationships between project information. There is an opportunity to design and evaluate the use of visualization techniques to visually communicate relationships between project information. This paper discusses our research efforts to prototype and evaluate two visualization techniques highlight and overlay that visually relate project information. Introduction Today, AEC professionals produce project information in electronic form with discipline-specific tools. Much research has focused on developing methods to integrate this information and standardize how AEC data and their relationships are modeled [IAI 1998]. Large scale displays to view these information models are becoming economical. However, no tools provide functionality to visually communicate the relationships between project information. Consequently teams must spend a lot of time mentally relating project information to support decisionmaking tasks [Fischer et al. 2000]. Visualization techniques that visually communicate the relationships between project information can potentially improve a teams ability to relate project information and improve the overall decision-making process. Consider the following hypothetical scenario in which a project team uses a CIW with two visualization techniques -highlight and overlay to review a project

Visualization of construction planning information

A construction schedule helps planners to identity potential construction problems prior to actual building construction. Planners must envision the sequence of construction, the workspace logistics, and utilization of resources and equipment in space and over time. This paper discusses methods of generating, visuaMng, and evahrating construction planning information with CAD based tools. We use a construction example to illustrate how feature extraction of 3D CAD models can help identify construction problems and evaluate the quality of a construction plan through 4D analysis and 4D annotation.

A performance-based approach to wheelchair accessible route analysis

Abstract This paper presents a method to determine if a usable wheelchair accessible route in a facility exists using motion-planning techniques. We use a ‘performance-based’ approach to predict the performance of a facility design against requirements of a building code. This approach has advantages over the traditional ‘prescriptive’ code-based approach for assessing acceptability of designs, which is normal practice today for assessing wheelchair accessibility. The prescriptive method can be ambiguous, contradictory, complex, and unduly restrictive in practice, and it can be ad hoc and difficult to implement as a computer application. The performance-based approach directly models the actual possible behaviors of an artifact (in this case, wheelchair motion) that are related to the functional intent of the designed system (a building) and (hopefully) to the specification of a prescriptive building code. This paper presents example cases from architectural practice to illustrate the use of robot motion-planning techniques for wheelchair accessibility analysis. This application is an example of using modern computational methods in support of knowledge-intensive engineering. The simulation method has broad applicability within engineering design. We illustrate and discuss how to analyze virtual simulations of the detailed behavior of a designed artifact in order to assess its use by intended users.

Requirements and Benefits of Interactive Information Workspaces in Construction

This paper envisions a new type of construction information technology- construction information workspace (CIW) — that enables project teams to visually interact with project information to support the decision-making process. We introduce a set of visualization and interaction functional requirements for CIW that will provide teams with the ability to visually interact with relationships between project information. These requirements stem from three research efforts: observation of todays paper-based workspaces, observation of a 4D workspace on a construction project, and prototyping of an interactive information workspace. These observations show the potential opportunities and benefits for the use of CIW such as improving the utility of project information and improving decision-making.

Rapid conceptual design evaluation using a virtual product model

Abstract This paper presents an architecture and test results for a computer-based system for assisting the conceptual phase of building design. The system uses 3D CAD to represent a graphic model of the design, and it uses AI symbolic models of the geometric forms, intended functions and computed and assigned behaviors of the design. The system uses A1 symbolic reasoning methods to analyze design behavior and compare predicted behavior with intended function. The Semantic Modeling Extension (SME) system incorporates a virtual product model: a small but extendible set of classes that define generic forms, functions and behaviors of facilities. After drawing a design using 3D CAD, a designer interactively creates interpretation objects as instances of the virtual product model. The interpretation objects express the meaning of the graphic representation with respect to a particular engineering issue, such as energy use or cost. The interpretation represents geometric and topological attributes of the features for use by automated design analysis tools. Interpretation objects unite support for graphically-oriented design thinking with support for automated symbolic reasoning. The paper includes an example building design scenario using the software prototype, illustrating how interpretation of the geometric model produces a symbolic model and supports multiple and changing analyses and evaluations during design. Students and practising engineers have tested the system in classes and workshops.

The Virtual Design Team: A Computer Simulation Framework for Studying Organizational Aspects of Concurrent Design:

Successful concurrent design requires a well organized design team. Without analysis tools for organization design, managers must rely on their experience or trial-and-error, rather on systematic generation and evaluation of alternatives, to design their organizations. The goal of the Virtual Design Team (VDT) research project is to develop computerized analysis tools to support the sys tema tic design of organ iza tion structures for complex, project-oriented tasks. The Virtual Design Team is a computational discrete event simulation model that incorporates qualitative reasoning concepts derived from Artificial Intelligence research. VDT explicitly incorporates information processing and communication models from organization theory and allows qualitative predictions of organizational performance. VDTs behavior has been validated extensively for internal consistency. We have also validated that its behavior compares well with theoretical predictions about, and the observed behavior of, real design project teams for petrochemical refinery, offshore oil systems, and power plant construction projects.