Professional Documents
Culture Documents
01 A Novel Approach To Expert Systems For Design of Large Structures
01 A Novel Approach To Expert Systems For Design of Large Structures
01 A Novel Approach To Expert Systems For Design of Large Structures
54 AI MAGAZINE
for Design of Large Structures
WINTER 1988 55
ened cylindrical composite panels and
shells. The knowledge base of this
expert system, called the buckling
OPTIMUM LAYOUT TRUSS TYPE = PARKER expert, contains knowledge of various
MEMBER NUMBERING SPAN LENGTH (FT.) = 230.0
FT. HEIGHT (FT.) = 33.0 analysis methods, when and how to
NUMBER OF PANELS = 10 use them, and how to interpret the
results. The knowledge in the system
14 15 16 17 was acquired from a journal article
13 42 41 18
20.0 12
43 40
39 19
that summarizes the experience of an
44
11 45 22 23 24 25 26 27 28 38 20
expert in the field.
21 30 31 32 33 34 35 36 29 Chehayeb, Connor, and Slater
0.0 37
1 2 3 4 5 6 7 8 9 10 (1985) report the development of a
general engineering problem solving
environment (GEPSE), in C. The
0.0 50.0 100.0 150.0 200.0 choice of the C language was based on
FT.
its transportability and efficient
numeric processing. Engineering
knowledge is divided into static and
active knowledge. Static knowledge
Figure 1. An example of a Parker Truss with includes the physical description of an
Member Numbering Plotted by BTExpert. engineering system. Active knowl-
edge is defined as scientific laws and
methodology to the development of design is presented by Maher, Fenves, heuristic rules that must be satisfied
an expert system for mechanical and Garrett (1988) in a recent article, in a particular domain. GEPSE has
design with design refinement as the using a Hearsay-II–like (Erman et al. been used for description and
central problem-solving activity. 1980) blackboard architecture. Sriram verification of a simply supported
One of the widely cited expert sys- (1986) presents a conceptual model for reinforced concrete beam subjected to
tems for structural design is Hi-Rise, integrated structural design called a uniformly distributed load.
developed by Maher and Fenves (1985) Destiny. Destiny is also based on a Adeli and Al-Rijleh (1987) present a
at Carnegie-Mellon University. Imple- blackboard architecture. The knowl- coupled expert system—the roof truss
mented in production schema repre- edge base of Destiny has a three-level expert (RTExpert)—for the design of
sentation language (PSRL), Hi-Rise is hierarchical structure. roof trusses. RTExpert can advise the
a knowledge-based system for prelimi- Composites design assistant (CDA), user on the appropriate type of the
nary design of rectangular commercial an expert system for design of a sand- roof truss, selection of the layout of
or residential high-rise buildings wich panel made of a honeycomb or the truss (such as the pitch of the
—those more than 10 stories high. foam core bonded to metallic or com- truss and the number of panels), and
The selection of a structural system posite face sheets, was developed by the loading. The design basis is the
in actual practice is usually based on a Pecora, Zumsteg, and Crossman AISC specification (American Insti-
variety of factors, including aesthet- (1985). CDA consists of a backward- tute of Steel Construction 1980). The
ics, economics, efficiency, and struc- chaining expert system shell written truss is designed for dead, live, snow,
tural integrity. Hi-Rise uses weighing in Prolog, a relational database man- and wind loads in accordance with the
factors in a linear-evaluation factor to ager written in Fortran, a laminate American National Standards Insti-
evaluate the merits of different struc- analysis program also written in For- tute (ANSI) specification (American
tural systems. Hi-Rise selects two tran, and a rule-based knowledge base. National Standards Institute 1982).
functional systems, that is, lateral Honeycomb core material, as well as A novel part of RTExpert is the
(wind or earthquake) load and gravity various metallic and composite mate- automatic computation of nodal
load resisting systems. Hi-Rise pre- rial properties, is obtained from the forces due to various loads. The user
sents all structurally feasible systems relational database manager. The anal- needs to indicate only the types of
as well as the best design according to ysis program can take into account materials used as roof covering and
the criterion of the linear-evaluation the hygrothermal effects, mechanical the location of the structure in the
function. The Hi-Rise knowledge base loading, viscoelasticity, and various United States. RTExpert automatical-
is obtained from textbooks. failure modes. The knowledge base in ly generates all the nodal forces. The
Sriram, Maher, and Fenves (1985) CDA was acquired from two compos- knowledge base and explanation facil-
present a small knowledge-based sys- ites handbooks. CDA interacts with ity of RTExpert were developed using
tem for checking structural steel the user iteratively through a se- the Insight2+ expert system shell. The
members for compliance with the quence of menus in order to produce a mathematical computations, graphic
American Institute of Steel Construc- satisfactory design. algorithms, and data-file manipula-
tion (AISC) specification (American Zumsteg and flaggs (1985) describe tion routines were developed in Turbo
Institute of Steel Construction 1980). a proof-in-concept system to be used Pascal. RTExpert has a comprehensive
A framework for detailed structural during the preliminary design of stiff-
56 AI MAGAZINE
graphic interface for displaying the
truss configuration, cross sections,
loading, and deformed shape. Informa- AASHTO LIVE LOAD
tion about individual members is pre- CLASS = HS 20
18.0 KIPS
sented through multiwindow graph-
ics-text displays. 0.64 KIPS/FT. OF VARIABLE LENGTH
Paek and Adeli (1988a, 1988b) devel-
oped a structural design language (SDL)
in the Interlisp environment for build-
ing coupled knowledge-based expert 1 UNIFORM LANE LOADING
systems for the integrated design of
structures. The complex body of knowl- 8.0 KIPS 32.0 KIPS 32.0 KIPS
edge needed for the detailed design of a 14 FT. V FT.
structure is fractionated into smaller
and manageable knowledge sources that
are organized into a hierarchy of cooper- 2 TRUCK LOADING
ating conceptual specialists. SDL has
NOTE: 1. All concentrated loads are in KIPS (=1000 lbs)
been used to develop an expert system, 2. Uniformly distributed loads are in KIPS/FT.
called Steel Design Expert (Steelex), for 3. V=Variable Spacing (14-30)
the integrated design of steel building PRESS PFKEY TO CONTINUE
structures consisting of moment-resist-
ing frames. Steelex designs the beams Figure 2. Example of an AASHTO
and columns making the frame as well Live (moving) Load Generated by BTExpert.
as the moment-resisting connections.
Steelex has a multiwindow graphics expert systems for design problems, a three categories: two-axle truck (H 15
interface that can display orthographic prototype expert system—the bridge and H 20), two-axle truck plus one-
and isometric views of the structure truss expert (BTExpert)—was developed axle semitrailer (HS 15 and HS 20),
and moment-resisting connections. for the optimum (minimum weight) and uniform lane loadings consisting
design of bridge trusses. The scope of of a distributed load of uniform inten-
A New Approach to Expert BTExpert is currently limited to the sity but variable length and a single,
optimum design of four types of bridge moving concentrated load (figure 2).
Systems for Structural Design trusses, that is, the Pratt, Parker, paral- The process of finding the maximum
The fundamental method of knowl- lel-chord K trusses, and the curved- forces as a result of live loads acting on
edge acquisition recommended in chord K truss for a span range of a bridge structure is not straightforward
practically all the recent books on 100–500 feet. An example of a Parker because of the complexity of AASHTO
expert systems is to find one or sever- truss with member numbering as plot- live loads. A heuristic approach was
al human experts in the problem ted by BTExpert is shown in figure 1. developed for finding the maximum
domain and use their knowledge in Design constraints and the moving compressive and tensile forces in the
the expert system (Hayes-Roth, loads acting on the bridge are based on members of a bridge truss based on the
Waterman, and Lenat 1983; Weiss and the American Association of State shape classification of the influence
Kulikowski 1984; Waterman 1986). In Highway and Transportation Officials line diagrams (ILDs) and the type of
fact, this approach was used in devel- (AASHTO) specifications (American AASHTO live loads (Adeli and Balasub-
oping the most celebrated expert sys- Association of State Highway and ramanyam 1987a). ILD is defined as a
tems in the fields of medical diagnosis Transportation Officials 1983). The figure showing the variation of some
(for example, Mycin), mineral explo- design of such a structure is highly behavior functions of the structure
ration (Prospector), and computer complicated in part because of the com- (axial forces in the case of trusses)
configuration (XCON). plex nature of AASHTO moving loads. when a unit load moves across the
Our approach to expert systems for structure. ILDs are used to find the
design in general, and structural A Heuristic Approach maximum forces in the truss members
design in particular, is novel in at for the Analysis of Bridges as the load moves across the bridge.
least two respects. First, mathemati- The heuristic procedure uses infor-
cal optimization is introduced in the
under Moving Loads
mation about the shape of ILDs for
expert system. Second, we use the Bridges are to be designed for com- the bridge truss type under considera-
machine to obtain parts of the knowl- bined dead and live (moving) tion. For statically indeter minate
edge necessary in the expert systems. loads.(Bridges must be designed for trusses, this information is obtained
We are, thus, extending the current moving loads such as the one shown through machine experimentation for
prevalent concept of expert systems in figure 2). Live loads are usually any given type of truss. The ILDs for
by incorporating machine intelligence specified by design specifications. member axial forces of a bridge truss
into the expert system. AASHTO live loads are used in BTEx- are classified according to their
As an example of our approach to pert. These loads can be classified into shapes. For Pratt trusses, for example,
WINTER 1988 57
T = Tension design, a hybrid optimization algo-
C = Compression rithm was developed for minimum
YT1 weight design of bridge trusses sub-
YT2 jected to moving loads (Adeli and Bal-
YT4
YC2
asubramanyam 1988b). In this algo-
rithm, an efficient zero-order explicit
approximation is combined with a
14 ’ 14 ’ YT3 more accurate but less efficient
YC3 YC5
’
YC1 ”
YC1 explicit stress-constraint formulation.
First, optimization is performed
14 ’
using the zero-order explicit approxi-
mation until the objective function
14 ’ 14 ’ 14 ’
attains a stationary value; then, the
control is transferred to the explicit
stress-constraint formulation. This
transfer control is performed automat-
CASE 1 YC1
’ Y C1
” ically by BTExpert using heuristic
rules. Note that BTExpert finds the
optimum detailed design of relatively
large structures subjected to the non-
linear and discontinuous constraints
YT1 of the AASHTO specification. This
YT2 optimization process requires sub-
YT4 stantial central processing unit (CPU)
YC4
time. The hybrid algorithm was devel-
oped in order to minimize the CPU
14 ’ 14 ’ ’
YC1 YC5
YC3 time for the mathematical optimiza-
YC2 ”
YC1 tion process.
14 ’ Architecture of BTExpert
BTExpert was developed using the
expert system development environ-
14 ’ 14 ’ 14 ’ 14 ’
ment (ESDE) (IBM Corporation 1986b,
1986c) and the expert system consulta-
tion environment (ESCE) (IBM Corpora-
CASE 2 YC1
’ < YC1
” tion 1986a, 1986b) implemented in Pas-
cal/VS. The first program is used to
develop expert systems and, in particu-
Figure 3. General Shape of Influence Line
lar, the knowledge bases. The second
Diagram Type 2 with Its Characteristic Ordinates.
program provides facilities for execut-
ing them. The two programs are collec-
a careful examination of some 5400 Kips load in figure 2 {1 Kip = 1000 tively referred to as the expert system
ILDs generated by computer for vari- pounds]) and W2 (the 8 Kips load in environment (ESE). The analysis and
ous layouts and truss member sizes figure 2) are needed. The decision tree optimization algorithms were coded in
guided us to conclude that all ILDs for finding these ordinates is present- Fortran 77. A schematic representation
can be classified into four types. Fig- ed in figure 4. This heuristic proce- of the architecture of BTExpert is
ure 3 shows ILD type 2 as an example. dure, which is based on the pattern shown in figure 5. The various compo-
(ILD is a diagram showing the varia- recognition of ILDs, results in sub- nents of BTExpert are briefly described
tion of the axial force in a truss mem- stantial savings in structural analysis in the following subsections.
ber when a unit vertical load moves computations. For details of this pro- (Most components of BTExpert are
across the structure.) cedure, see Adeli and Balasubra- described in the text. [Note the box
Decision trees and heuristic rules manyam (1987a). marked truss geometry in figure 5.]
were developed for finding the maxi- The truss geometry is generated auto-
mum compressive and tensile forces Mathematical Optimization matically [the user does not have to
in the members of a given truss type. The optimum design of a bridge truss input the coordinates of the truss
As an example, in order to find the consists of selecting the right combi- joints]. Display algorithms are devel-
maximum tensile forces in the mem- nation of the cross-sectional areas of oped for displaying the AASHTO live
bers of a Pratt truss whose ILD is type the truss members to satisfy all the loads [for example, figure 2], the truss
2, the ILD ordinates corresponding to design constraints and produce a least configuration with node or member
the location of wheel loads W1 (the 32 weight truss. To create the optimum numbering [for example, figure 1],
58 AI MAGAZINE
ILDs, and the convergence history of
the objective function and design vari-
COMPUTE YT1 , YT2
ables. The W-sections database con-
tains the cross-sectional properties of
all the 187 wide-flange steel shapes
commonly used in steel structures
and given in the AISC manual.) NO
LOAD CLASS=HS
User interface is provided in the
form of visual edit screens and menus
YES
in which the user types the values of
the required parameters into the
appropriate field. Further, using the YT2 = YT2 YT1 = YT2
graphical data display manager
(GDDM) (IBM Corporation 1984), a
graphic interface was developed for COMPUTE YT3 ,YT4
displaying the truss configuration
with joint or member numbering
(figure 1, for example), the design
AASHTO live loads (figure 2, for NO
YES
example), and ILDs for various mem- YT3 0
ber axial forces (figure 3, for example).
The explanation facility helps the
YES NO
user to examine the reasoning process. YT3 YT4 YT4 0
The explanation consists of both the
RULE text and RULE comments coded
by the knowledge base builder. The NO YES
explanation facility commands follow:
YT2 = YT3 YT2 = YT4 YT2 = YT4 YT2 = 0
1. EXHIBIT—It displays the current
value(s) of a specific parameter.
2. HOW—It displays an explanation of
how the system determined a value
for a parameter. An example of the
explanation generated by BTExpert in Figure 4. Decision Tree for finding the Maximum Tensile Forces YT 1and YT 2
response to the HOW command dur- Corresponding to the Location of the Wheel Loads W1 (the 32 Kips load in
ing a sample consultation is given in figure 2) and W2 (the 8 Kips load in figure 2). 1 Kip = 1000 lbs.
figure 6.
3. Why—It displays an explanation of
why the system is asking a given
DEBUGGING
question. An example of the explana- USER
FACILITY
tion generated by BTExpert in
response to the WHY command dur- USER INTERFACE
EXPLANATION
FACILITY
ing a sample consultation is presented TEXT BOOKS
in figure 7. DESIGN MANUALS
WORKING
4. What—It displays more informa- KNOWLEDGE BASE MEMORY
KNOWLEDGE BASE
tion about a given parameter. RESEARCH PAPERS
BUILDER DOMAIN KNOWLEDGE
Knowledge Acquisition. In BTExpert,
MACHINE CONTROL KNOWLEDGE
domain knowledge is obtained in part EXPERIMENTATION
from textbooks, design manuals,
design specifications (for example, RESULTS FROM
NEW DESIGNS
American Association of State High-
way and Transportation Officials INFERENCE MECHANISM
PROCEDURAL
1983), and research papers and journal INTERFACE FORWARD CHAINING
articles. In addition to these sources, to BACKWARD CHAINING
bridge the gaps in the knowledge base,
a detailed numeric machine experi-
mentation in the problem domain was TRUSS DISPLAY ANALYSIS OPTIMIZATION W-SECTIONS
undertaken to obtain the optimum val- GEOMETRY ALGORITHMS ALGORITHMS ALGORITHMS DATABASE
WINTER 1988 59
areas of truss members for various span
Focus: FCB13
lengths, AASHTO live loads, and grades
— How —
of steel (figure 8). The information
I assigned value to TERMINATE DESIGN of GLOBAL by
PF1 Help obtained through the machine experi-
1. The default constraint: assigned = ‘False’ (1). PF2 Review mentation was subsequently used in
2. Rule RULE0141 of GLOBAL (4) which states that PF3 End the knowledge base of BTExpert.
PF4 What The optimum layout optimization
If Optimization formulation is ‘Explicit Stress Constraints’
and Iteration number > Maximum number of zero order explicit approxima- PF5 Question parameters (that is the optimum
tion iterations + 1 PF6 Unknown height of the bridge and optimum
and (iteration number > (Maximum number of zero order explicit approxima- PF7 Up number of bridge panels, as well as
tion iterations + PF8 Down initial approximate cross-sectional
Maximum number of explicit stress constraints
formulation iterations) or Percentage change in the value
PF9 Tab areas) are obtained through interactive
of the objective funtion PF10 How numeric machine experimentation.
<Tolerance on percent change in the objective function) PF11 Why For each type of truss, this knowledge
Then TERMINATE DESIGN is ‘True’. PF12 Command is obtained for different AASHTO
This rule is a test for stopping the iterative cycle of analy- loadings and various types of steel.
sis/optimization of the indeterminate bridge trusses when Knowledge Base. The domain knowl-
the displacement constraint is not active. edge of BTExpert is represented in the
form of parameters and rules, and the
As a rusult of this rule control knowledge is represented in
TERMINATE DESIGN assigned = ‘True’ (1). the for m of focus control blocks
(FCBs). Each FCB can own some
Figure 6. An Example of the Explanation Generated by BTExpert in Response
parameter(s) or rules of the knowledge
to HOW It Arrived at the Value of the Parameter Terminate_design.
base. Because FCBs are the driving
mechanism for problem solving, each
parameter and rule should be refer-
enced in some FCB. If a parameter is
Focus: FCB2 4 associated with multiple FCBs in a
— Why — hierarchy, each association is treated
I am asking about Bridge loaction of FCB2 as a separate instance. The control
PF1 Help
to find Recommended AASHTO live load of FCB2 which I knowledge of BTExpert is classified
PF2 Review
am trying to determine. into 17 FCBs. The main idea of using
PF3 End
These rules are used for this line of reasoning. PF4 What FCBs is to express the complex opti-
PF5 Question mum design process into distinct
— RULE RULE0014 — PF6 Unknown steps and identify the intended use
If Bridge location is ‘Interstate’ PF7 Up and application sequence of rules and
Then Recommended AASHTO live load = ‘HS 20’ PF8 Down external procedures.
(Choose one of the following) PF9 Tab As an example, FCB1 owns the
PF10 How rules for selecting the right type of
— State road truss for a span length inputted by the
PF11 Why
— Trunk highway user. A sample rule in this FCB is as
PF12 Command
— Interstate follows: If Span_length > 300 and
— Main highway Span_length <= 400 Then Recom-
— Arterial highway mended_truss_type is `Parallel-chord
K truss’
Figure 7. An Example of the Explanation Generated by BTExpert in Response FCB2 contains the rules for select-
to WHY It Is Asking the Value of the Parameter Bridge_location. ing the right type of design live loads
for the bridge under consideration. A
tion, a software for layout optimiza- contrast to formal shape-optimization sample rule in this FCB is as follows:
tion of trusses, called Interactive Opti- procedures that are computationally If Bridge_location is `State road’ and
mization of Trusses (IOTRUSS), was expensive and can produce impractical Traffic_intensity is `Light’ Then
developed in Fortran 77 (Adeli and Bal- designs, the synergic man-machine AASHTO_live_load = `H 15’
asubramanyam 1987b). The layout approach used in IOTRUSS is an effec- The FCB3 contains the rules for obtain-
optimization in IOTRUSS is based on tive method for practical layout opti- ing the yield stress and the relative cost
changing certain key dimension(s) of mization of trusses. of the steel used in the truss bridge. If
the truss and performing optimization The software IOTRUSS was used as a Steel_type is `M 183’ Then Yield_stress
for each layout by taking advantage of knowledge-acquisition tool to find the = 36 and Relative_cost = 1.0
the interactive environment of the optimum values for the height, number Another FCB calculates the thickness
computers with graphic facilities. In of panels, and initial cross-sectional of the deck slab, and so on (Adeli and
60 AI MAGAZINE
Balasubramanyam 1988a).
Procedural interface. For numeric
BRIDGE TRUSS
processing and graphics interface, SPAN 100 & 500 FT.
BTExpert uses procedures implement-
ed in Fortran 77. Therefore, an inter-
SPAN 100 & 200 FT. SPAN > 200 & 300 FT. SPAN > 300 & 400 FT. SPAN > 400 & 500 FT.
face was developed in PASCAL/VS PRATT TRUSS PARKER TRUSS PARALLEL-CHORD K TRUSS CURVED-CHORD K TRUSS
WINTER 1988 61
readily available or acquired from BTExpert can be modified and extend- Specifications for Highway Bridges, 13th ed.
human experts. Consequently, the ed to incorporate the cost of the struc- Washington, D.C.: American Association of
performance of such systems can be ture instead of the weight. However, State Highway and Transportation Officials.
compared to human experts. The mul- this modification is more a develop- American Institute of Steel Construction
tifacet research presented briefly in ment issue than a research issue. 1980. Manual of Steel Construction. Chica-
this article, however, attacks a prob- BTExpert is currently being extend- go: American Institute of Steel Construc-
lem for which a single human expert ed. Heuristic rules and procedures are tion.
does not exist. Thus, the performance being developed to improve the American National Standards Institute
of BTExpert cannot be compared to efficiency, robustness, and accuracy of 1982. American National Standard Mini-
human experts. In other words, in a the optimization process. For example, mum Design Loads for Buildings and Other
sense, BTExpert outperforms human heuristic rules are being developed for Structures. New York: American National
experts because it uses mathematical the choice of the right optimization Standards Institute.
optimization techniques and the algorithm and appropriate control Bennett, J. S., and Engelmore, R. S. 1979.
knowledge obtained using interactive parameters using machine learning. SACON: A Knowledge-Based Consultant for
machine experimentation. Other heuristic rules are being devel- Structural Analysis. In Proceedings of the
BTExpert addresses a realistic oped for the classification of con- Sixth International Joint Conference on
design problem. Therefore, a question straints into inactive, partially active, Artificial Intelligence, 47–49. Menlo Park,
arises about why no single human active, and violated constraints. The Calif.: International Joint Conferences on
expert exists to solve the bridge inactive constraints will not be includ- Artificial Intelligence.
design problem. Of course, there are ed in the optimization cycle through a Brown, D. C., and Chandrasekaran, B. 1984.
many bridge designers who can design constraint deletion process. This strate- Expert Systems for a Class of Mechanical
truss bridges. For given design and gy will result in a more efficient opti- Design Activity. In Proceedings of the Inter-
national Federation for Information Pro-
loading conditions, each bridge mization algorithm. Finally, a sample
cessing WG5.2 Working Conference on
designer will come up with a different consultation with BTExpert is present-
Knowledge Engineering in Computer-Aided
design, and most probably, no one will ed in a forthcoming book by Adeli and
Design.
come up with the optimum design. Balasubramanyam (1988b). The exam-
However, there are university ple presented here covers an entire Chehayeb, F. S.; Connor, J. J.; and Slater, J. H.
1985. An Environment for Building Engi-
researchers who are knowledgeable chapter of this book.
neering Knowledge Based Systems. In Appli-
about mathematical optimization
References cations of Knowledge-Based Systems to
algorithms. These researchers usually Engineering Analysis and Design AD-10, ed.
apply their optimization algorithms to Adeli, H., ed. 1988. Expert Systems in Con- C. L. Dym, 9–28. New York: American Soci-
the solution of academic problems. struction and Structural Engineering. Lon- ety of Mechanical Engineers.
The practicing engineers, however, are don: Chapman and Hall. Dixon, J. R., and Simmons, M. K. 1983.
not familiar with mathematical opti-
Adeli, H. 1986. Artificial Intelligence in Computers That Design: Expert Systems for
mization algorithms. In fact, the Mechanical Engineers. Computers in
Structural Engineering. Engineering Analy-
detailed optimization of realistic sis 3(3): 154–60. Mechanical Engineering, 10–18.
structures (such as bridge trusses) sub-
jected to realistic design constraints Adeli, H., and Al-Rijleh, M. M. 1987. A Elias, A. L. 1983. Computer-Aided Engineer-
Knowledge-Based Expert System for Design ing: The AI Connection. Astronautics and
(which are usually implicit and dis-
of Roof Trusses. Microcomputers in Civil Aeronautics, 48–54. American Institute of
continuous functions of design vari-
Engineering 2(3): 179–195. Aeronautics and Astronautics.
ables) is a highly nonlinear problem,
with the associated problems of Adeli, H., and Balasubramanyam, K. V. Erman, L. D.; Hayes-Roth, F.; Lesser, V. R.;
efficiency, convergence, and stability. 1988a. A Knowledge-Based System for and Reddy, D. R. 1980. The Hearsay-II
Design of Bridge Trusses. Journal of Com- Speech Understanding System: Integrating
These problems explain why practic-
puting in Civil Engineering 2(1): 1–20. Knowledge to Resolve Uncertainty. Com-
ing engineers shy away from mathe- puting Surveys 12(2): 213–253.
matical optimization algorithms. Adeli, H., and Balasubramanyam, K. V.
BTExpert attempts to bring mathe- 1988b. Expert Systems for Structural Hayes-Roth, F.; Waterman, D. A.; and Lenat,
matical optimization techniques to Design—A New Generation. Englewood D., eds. 1983. Building Expert Systems.
Cliffs, N.J.: Prentice-Hall. In press. Reading, Mass.: Addison-Wesley.
the reach of practicing engineers and
designers. Adeli, H., and Balasubramanyam, K. V. IBM Corporation 1986a. Expert System Con-
BTExpert is a prototype expert sys- 1987a. Heuristic Analysis of Bridge Trusses sultation Environment User Guide, SH20-
under AASHTO Live Loads. Microcomput- 9606-1, IBM Corporation, Cary, North Car-
tem developed in an academic envi-
ronment. We used the minimum ers in Civil Engineering 2(2): 147–160. olina.
weight as our optimization criterion. Adeli, H., and Balasubramanyam, K. V. IBM Corporation 1986c. Expert System Con-
A more realistic criterion would be 1987b. Interactive Layout Optimization of sultation Environment and Expert System
the minimum cost. In this case, the Trusses. Journal of Computing in Civil Engi- Development Environment Reference Man-
neering 1(3): 183–196. ual, SH20-9609-1, IBM Corporation, Cary,
labor cost must be added to the cost of
materials. However, for large steel North Carolina.
American Association of State Highway and
structures, the cost of materials (steel) Transportation Officials 1983. Standard IBM Corporation 1986c. Expert System
is the primary cost of the structure. Development Environment User Guide,
62 AI MAGAZINE
AAAJ is plcascd to announce publication of Ren&gs fto~n AI Mngnzine, the complete collection of all the
alticlcs that appeared during AI Magnzine'sfirsst five years. Within this &O-page indexed vohunc, you will
find articles on AI written by the foremost practitioners in the field-articles that catncd AI Ma@zinethe
title “journa1 of record for the artificial intelligence community.” This collection of classics from the ptc-
micr publication devoted to the entire field of artificial intclligcncc is availablc in one large, paperbound
desktop reference.
Subjects Include:
* Automatic P~ogranuning * Distributed Altificisl Inteiligcnce l chnlcs
Maher, M. L.; Fenves, S. J.; and Garrett, J. H. Sriram, D. 1986. DESTINY—A Model for
1988. Expert Systems for Structural Design. Integrated Structural Design. In Proceedings
In Expert Systems in Construction and of the Conference on Computer-Aided
Structural Engineering, ed. H. Adeli, 85–121. Design 86, 226–235.
New York: Chapman & Hall. Sriram, D.; Maher, M. L.; and Fenves, S. J.
1985. Knowledge-Based Expert Systems in
Paek, Y., and Adeli, H. 1988a. Representa-
WINTER 1988 63