Evaluation of progressive, innovative materials and components used in housing by the use of fuzzy-logic

Information technology for construction management and
building performance evaluation support:
- a North-American perspective -
K. Menzel, A. Mahdavi, V. Hartkopf, S. Lee
Center for Building Performance and Diagnostics 5000 Forbes Avenue, Pittsburgh, PA 15213-3890, USA ABSTRACT: We provide a critical overview of the current status of computational support for construction
management and building performance evaluation in North-America. This overview is based on the research
conducted in relation to the design and construction of the Intelligent Workplace (IW), Carnegie Mellon University,
Pittsburgh, Pennsylvanie focus on the evaluations of selected commercial and research software
products particularly in view of their relevance for interdisciplinary co-operation between planners and managers as
well as decision support for early stages of design.
INTRODUCTION
Before we focus on our specific topic, the use of information technology for construction management and building
performance evaluation, we would like to include some selected results of the latest survey focusing on the role of
computing conducted by the Education Committee of the Technical Council on Computer Practices of the American
Society of Civil Engineers presented and published in. One of the survey's goals was to identify the status of
computing skills of new hires (or new graduates). Based on this status, practitioners asked about their expectations,
and recommendations regarding specific skills that need to be improved.
Besides the desire to have all graduates be skilled in the application of CAD-packages, most of the responses
contained the even stronger desire to have recent graduates consider the computer and the appropriate software
packages as a problem analysis tool instead of a problem solving tool. Most of the respondents felt that new
graduates are literate, but several statements also indicated concerns that recent graduates are too dependent on
computers and application software. A considerable number of respondents expressed that the need for new
engineers to develop stronger verification skills and a better understanding of the limitations of computers and the
software packages. As application software becomes more complex and sophisticated, so are the demands on the
development of certain skills such as logical approach to problem solving, debugging skills, development of
mathematical algorithms and a general understanding of how computers and communication infrastructure work.
We would like to use these introductory comments to state that our evaluation criteria for information technologies
in building management and building performance evaluation focus on the support which architects/engineers can
gain/expect from complex software tools during the analysis of products and processes.
CONSTRUCTION MANAGEMENT
The focus of the following section is directed towards research trends in the development of construction
management software packages that may be regarded as representative within the United States of America and the
coverage of typical construction management software products available in the U.S. market, since a completeoverview is neither feasible nor meaningful within the scope of this paper.
Construction Planning, Scheduling and Resource Allocation
SOFTWARE TOOLS: The commercially available software packages are distinguished within two categories (1)
high end packages and (2) mid range packages. The high-end packages are built on a standard data-base, have
network capabilities and have larger capacity. During the mid-nineties some reviews and comparative evaluations
were published (detailed informati). Some specific characteristics of these tools are briefly discussed in
more details in the following paragraphs.
The implementation of data entry and the representation are in most of the cases graphic-based. Various types of
charts (Gantt charts, PERT charts) resource histograms, task lists and calendars are available. The data representation
in various views enable the necessary flexibility. Data can be edited either in the graphical representations or within
the tabulated forms.
Project scheduling tasks are in most packages managed by using the Critical Path Method (CPM). Most of the high-
end packages will allow a version management. Some of the packages can disassemble a project into subprojects to
be distributed to individual managers. Changes of the subprojects are automatically inserted into the master plan
when the subprojects are reassembled.
Resource histograms are available in most of the packages as well as automatic leveling functions to reallocate
resources. The user can control the percentage of completion of construction by time, cost, and work in most of the
commercially available software tools. Various reports can be generated easily. A detailed cost monitoring is
possible in most of the packages because features like incorporated inflation factors, overtime cost rates, and
comparison of actual versus budget costs are available. The level of networking capabilities varies. Selective features
such as file locking or multi-level password protection are not available in every package. File based data exchange
is available within all of these packages. Some of the high-end packages offer superior integration with other
software.
RESEARCH: There are two major research foci in general:
(1) Efforts to facilitate or automate specific project management tasks more efficiently, and(2) Efforts to develop integrated computer systems that allow effective communication among the various specific planning tasks as well as the cooperation with other planning and design activities.
Researchers at the University of Michigan developed a case based planning and scheduling system. Their researchactivities focus on the reusability of contractors' experiences from repetitious and similar projects for the planning ofnew facilities. A general product model is used to model facilities and organize planning options.
Other research activities in the early nineties focused on AI-application, rule-based expert systems, and knowledgebased expert systems. More detailed information is available in pp. 92 to 94.
Cost Estimating
SOFTWARE TOOLS: Many software tools can generate detailed cost estimates by using automatic quantity take-off
process with CAD-systems. Material databases are available for users and can be modified by users to specify
building materials associated with building elements. Some tools can act as an integrator, collecting relevant design
and quantity data from various CAD-systems and integrate these data into a final cost estimate.
Few software packages have incorporated extensive decision support functionalities. The decision support is
available through historical cost data based on existing buildings and the capability to extend the existing database
with cost data from self designed projects.
Construction Simulation and Visualization
SOFTWARE TOOLS: The number of commercially available software packages with construction simulation and
visualization capabilities are limited.
RESEARCH: Most of the methods that attempt to describe / simulate construction processes lack the capability to
incorporate the geometric component of building projects. The abstract representation of complex building cycles in
Gantt charts or CPM schedules represent the ability to communicate a progression of construction activities over a
certain time. But there still exists a need for the development of software packages which allow architects / engineers
and contractors / subcontractors to simulate and visualize construction sequences. The ongoing research effort at the
Center for Integrated Facility Engineering (CIFE) at Stanford University is focusing on this problem. Researchers at
CIFE develop a so called 4D-CAD system. A 4D-CAD represents time and space. It can be considered as a graphic
simulation of a process. The simulation involves the combination of a 3D-graphic model and a construction
schedule. The objective of the visualization part of the CIFE 4D CAD system is not to build a 4D-CAD VR tool
Researchers at CIFE are focusing on the development of a tool which simulates the various stages of a construction
process and to visualize these stages.
Research at the School of Civil and Environmental Engineering at the Georgia Institute of Technology (GATech),
Atlanta focuses on the VR-aspect. The research approach at GATech intends to place designers, construction
managers and other users into a virtual scenario. Researchers at GATech develop an object oriented virtual
environment which provides the user with several choices of virtual construction equipment. The virtual construction
equipment is able to receive instructions and carries out tasks needed to assemble buildings. It is controlled by
reactive control algorithms and thus able to adapt changes in the environment. Instead of finding ways to model the
system so the simulation runs correctly, the user can interact with the system.
The system model consists of six basic objects: camera, light, terrain, machine, building, and building object.
Building objects are the basic elements for a built object. These building objects are arranged in an hierarchical or
sequential fashion. The hierarchy is determined by the relationships that exist between these objects. The sequences
represent the order in which the building objects have to be assembled.
Each building describes an object that requires an assembly process. The assembly sequence is contained in the
building object definition file and can be extracted by the planner. She / he is able to distribute the contained task
descriptions to the appropriate machine.
Machines are composed of objects. These objects interact with each other in a hierarchical fashion. The capability of
interaction is achieved by the implementation of behavior-modules. The terrain object serves as physical support for
all components (building objects and machines).
Camera and light are necessary accessories for the graphical representation / visualization.
BUILDING PERFORMANCE SIMULATION AND EVALUATION
In the beginning we discuss problems of computational building performance and simulation tools. In this context
building performance stands for the evaluation of the behavior of buildings and their systems regarding energy use
and the thermal environment, as well as the control of natural and artificial lighting, acoustics and air quality. Issues
of building performance are usually covered within the field of "Bauphysik" and "Klimatechnik" in German-
speaking countries. Next, a brief review of the current research and the state-of-the art tools is given. The focus of
the following section is directed towards research trends that may be regarded as representative within the United
States of America and the coverage of typical products available in the U.S. market, since a complete overview is
neither feasible nor meaningful within the scope of this paper.
Energy and Buildings
The topic of energy and building design could be discussed from several perspectives, such as HVAC-system
design, impacts of enclosure and lighting systems, peak loads, and energy conservation. The focus of this part is
sizing of HVAC-equipment by the HVAC-designer and the analysis of energy consumption, because the current use
of the energy-related software tools is dominated by these topics.
SOFTWARE TOOLS: The numerous tools available in the U.S. market can be characterized as follows:
- The thermal process model of most of the available tools is incomplete. Although the more sophisticated tools
consider transient conduction, short-wave radiation, and air-exchange they typically do not take into effect long-wave radiation, convection, air-flow, detailed thermal interaction between building and the surroundingenvironment.
- There are various building models, specifying various levels of resolution, implemented within these tools. The more sophisticated tools use for example a multi-layer description to model walls which allows for modelingtransient conduction and thermal bridging.
- The accuracy of HVAC-system and plant modeling varies from purely empirical to very detailed models. The empirical models describe the HVAC-equipment as single units with certain operating characteristics, andtypically contain no information on physical dimensions, fluid properties, flow rates, pressures, temperatures, etc.
The detailed models are component based and employ more specific information.
- The external environment is modeled to various levels of detail. For example temperature effects can be characterized by the degree-day method or by the more sophisticated hourly weather data files.
RESEARCH There are two major research foci in general: (1) Efforts to improve the accuracy and reliability of the models and algorithms used for the calculations, and(2) Efforts to effectively integrate the use of simulation for design decision support.
Research in the first field includes:The SPARK object oriented model-lab environment, developed at the Lawrence Berkeley Laboratory, is intended toprovide the ability to build detailed simulation tools using low-level objects that encapsulate fundamental heat andmass transfer equations.
The use of neural networks to monitor and accurately predict energy consumption in existing buildings isdemonstrated in the research of the University of Colorado, Boulder.
Last but not least, we would like to mention that there are also efforts to enhance existing tools. For example DOE-2is regularly updated with more accurate ways of simulating various building and HVAC-system components.
A number of research efforts have been made to integrate the use of simulation for design and decision support. Alarge project, that involves Pacific Northwest Laboratory, Lawrence Berkeley Laboratory, California PolytechnicInstitute, and the University of Oregon is the AEDOT-project (Advanced Energy Design and OperationTechnologies). AEDOT is intended to provide design advice and guidance to support qualitative and quantitativeaspects of iterative and interdisciplinary design, construction, and operation tasks of buildings.
Researchers at the Center for Building Performance and Diagnostics are developing SEMPER, an active multi-aspectcomputational tool integrating building performance simulation (energy, lighting, and acoustics) into computationaldesign systems. Specifically SEMPER seeks to meet the following requirements: (1) a methodology consistentperformance modeling approach through the entire building design and engineering process; (2) seamless anddynamic communication between the simulation model and the general building representation; (3) a "preference-based" performance-to-design mapping technology for active design support.
Natural and Electrical Lighting
This is in relationship to software capabilities and research efforts for analyzing daylighting, and electrical lighting.
We do not consider the design of indoor electrical lighting systems in this paper.
SOFTWARE TOOLS: Few commercially available software tools provide daylighting evaluation. Most of them
consider electrical lighting design and evaluation. The following evaluation of currently commercially available
software is based on a survey of the Illuminating Engineering Society of North America (IESNA).
Most of the tools predict illuminance in a sufficient way; however, they are limited in that they only give an average
value for the reflected component of light throughout the space. This is because of the use of the zonal cavity
method.
Some programs calculate luminance but few of them resolve these luminance finely enough to produce images.
The more advanced lighting simulation programs use ray-tracing and radiosity techniques. The incorporation of
daylight evaluation is rare in commercially available software products. Very few of them consider the integrated
effect of daylight and electrical lighting. As mentioned earlier, only some of the software tools are able to provide
plots of the illuminance of the space as output. Most of them only provide numerical output results.
Most of the tools typically cannot model internal partitions, light shelves, and complex geometries, such as curved
walls and ceilings, sloped ceilings, and sloped apertures. Lighting simulation tools are less used by building
designers but often used by lighting-design consultants. As a result the scope for integrated evaluation and decision
making is rather limited.
RESEARCH :
The research foci in this area are comparable with the general research foci within the energy related research.
(1) Efforts to improve the accuracy and reliability of the models and algorithms used for the calculations.
(2) Efforts to effectively integrate the use of simulation for design decision support.
The research in improving accuracy and reliability includes:The extension of the use of radiosity, which was basically developed in the area of computer graphics, for lightingevaluation. This includes the combination of radiosity and ray-tracing techniques to model both global diffuse andspecular components of light, and a progressive refinement approach for fast image generation. Work on these topicswas performed in the late eighties at Cornell University. During the same time researchers at Lawrence BerkeleyLaboratory worked on the detailed simulation of the solar-optical properties of multi layer fenestration systems.
In the early nineties research efforts focused on the development of more accurate daylight modeling, includingmore realistic modeling of the sky. Research was performed at the University of Berkeley, the University ofMichigan, and the State University of New York. As an alternative to the daylight factor method, the coefficient ofthe utilization method has been studied at the Pennsylvania State University.
Research activities to efficiently integrate the use of simulation tools for decision support includes:The development of "GESTALT" an daylighting simulation environment that allows for the simultaneousmodification and observation of changes in design and performance variables. The tool was developed at CarnegieMellon University (CMU), Pittsburgh, PA. Currently, researchers at CMU are attempting to develop tools thatsupport integrated design of daylighting with electrical lighting design.
Air Quality
Air quality issues have become important with the increase in Sick Building Syndrome and other building air quality
problems over the last decade. However, air quality modeling and simulation has not become a part of building
design practice to the extent that energy and lighting modeling have, because of its complexity and the expensive
computational demands. It is mostly still in the research domain.
SOFTWARE TOOLS and RESEARCH: An international survey of 50 tools analyzing infiltration and ventilation airflow distribution was performed at the Lawrence Berkeley Laboratory by Feustel [FEU '92]. In relation to softwaresystems integration the survey revealed that none of the examined tools have CAD input or graphical output. Few ofthese tools are coupled with pollution migration tools or thermal models. Feustel also notes that the development ofmulti-zone infiltration and ventilation models was very slow and that there is less difference between modelsdeveloped in the early seventies and those developed in the late eighties. Since the early nineties computational fluiddynamics (CFD) has been applied to the study of natural ventilation, air conditioning, and smoke movement.
However, until recently the success in the application of CFD models was largely dependent on the skills of theoperator in dealing with the definition of the geometry, and the specification of boundary conditions. The recentincrease in computational power of desk-top computing led to an increased use of CFD in commercially availablesoftware products, such as FLOVENT (U.S. distributor: Flomerics, Westborough, MA) Summary
In the domain of building performance simulation and decision support one can notice that the development and
application of computational tools is industry driven. As a result the concerns addressed by the tools are mainly
issues pertaining to the selection and sizing of systems and components rather than an integrated performance
evaluation. Consequently, these programs are rarely used by building designers, especially in the early design stages,
where the predictive capabilities of simulation tools could be of significant value. Although many research
institutions address the necessity for the integration of performance simulation within the overall design support
environments, most of the available performance simulation tools still remain mono-dimensional and isolated. This
fact reflects the sequential character of the building design, planning and construction process as well as the
fragmented nature of the building industry. Truly integrated design and analysis software systems are still not
available.
With regard to the commercial software products in the field of construction management the following limitations
can be identified: although project planning, cost estimating and construction simulation are supported, tasks like
bidding as well as site and material management have not received the same level of attention. Few project
management software packages are integrated in a total design support software system and the data entry still
remains a strenuous procedure. Most packages still represent a "static" view of data. Little analysis or evaluation
options are provided to support managerial decision making. Various research groups address the construction
planning and scheduling, construction contracting, site layout generation as well as the integration of these three
topics. Currently problems such as efficient material management and calculation of environmental and energy
responsive site management are insufficiently addressed within the ongoing research projects.
REFERENCES
[ 1 ]
R.J. O'Neil, R. M. Henry, T.A. Lenox, "Role of Computing: Practitioners' Perspective", in: Proceedings ofthe Third Conference on Computing in Civil Engineering, pp. 670 - 676, Anaheim, CA, 1996 V. Hartkopf, et al., "State of the Art in Computer Applications for Architecture, Engineering andConstruction in the USA", Center for Building Performance and Diagnostics / Carnegie Mellon University,Pittsburgh, PA, USA, 1994 K. Mc Kinnley, J. Kim, M. Fischer, C. Howard, "Interactive 4D-CAD", in: Proceedings of the ThirdConference on Computing in Civil Engineering, pp. 383-389, Anaheim, CA, 1996 W. Ribarsky, J. Bolter, A. Opdenbosch, R. van Teylingen, "Visualization and Analysis Using VirtualReality", in IEEE Computer Graphics and Applications, Vol. 14, No. 1, pp. 10-12, 1994 M. Phair, " Construction Workers are Game for Virtual Equipment", in: Engineering News-Record, March18, 1996, pp. 30 - 33, McGraw-Hill Companies, 1996

Source: http://zuse.ucc.ie/~Karsten/Publications/1997/03_071Paper.pdf