g
1999 AAAI Spring Symposium on AI in Equipment Maintenance Service &Support

Symposium

Dates, times, and placesWhy a symposium now? •  Who should attend?Format • 
Preliminary ScheduleWorking SessionProceedingsAbout AAAI
Home 
Description 
Organizing Committee 

Symposium 

Equipment Service 
Background 
Examples 

Artificial Intelligence 
Challenge problems 

Software and Data 
Existing software 
Useful datasets 

References 
FAQ 
Web sites 
Bibliography 

Contact 
Comments 
Subscriptions

Dates, times, and places

The 1999 AAAI Spring Symposium on Artificial Intelligence and Equipment Service will be held Monday through Wednesday, March 22-24, 1999 on the Stanford Campus in Palo Alto, California. The Symposium is part of the AAAI Spring Symposium Series, which includes a number of other symposia on a variety of topics. An informal reception will be held on Monday, March 22. A general plenary session featuring the highlights of each symposium will be presented on Tuesday, March 23. 

Attendance at the Symposium is by invitation or by open registration. Invitation is based on submission of an extended abstract or statement of interest. The submission deadline is past. Open registration will be available through AAAI to a limited number of participants in early September on a first-come first-served basis. 

Important upcoming dates: 

  • October 30, 1998 - Abstract Submission 
  • November 13, 1998 – Invitations issued based on submissions 
  • December 15, 1998 – Registration Information available through AAAI 
  • January 29, 1999 - Camera-ready papers are due
  • February 5, 1999 – Invited participants registration deadline 
  • February 26, 1999 - Final (open) registration deadline
  • March 22-24, 1999 – Symposium held in "History 303" 

Why a symposium now?

The symposium is trying to distinguish itself in several ways from traditional monitoring and diagnosis activities. First, it acknowledges that there is a recent thrust of using AI in service. Second, this thrust has been driven partly by the hardware developments which allow unattended monitoring. This results in many new questions from autonomous decision making to adaptation in changing environments. Third, various diagnostic tools which now have access to a large number of data will come to different conclusions which in turn have to be resolved. Lastly, the focus is on service related activities. 

Who should attend? 
The symposium will explore AI techniques that can be developed and deployed within a very short time frame. We expect these technologies to enable significnt advances in equipment service applications. Several communities will draw useful insights from the symposium: 

  • AI Researchers – Researchers will be introduced to a fertile new application area for AI technologies for segmentation, classification, prediction, and decision making. They will hear about the key technical challenges facing existing application users, and discuss what technical advances would be most useful. Researchers will also be able to guide and inform future research directions and funding programs. 
  • Potential users of equipment service – Users will hear about existing and emerging AI technologies with the ability to radically transform current equipment service applications. In addition, they will be able to pinpoint key technical problems for AI researchers to address. 
  • Software developers – Developers will hear assessments of the current state of the art for many relevant technologies, helping them to evaluate which technologies to incorporate into future products. In addition, they will be able to make important contacts in the research and applications communities. 
  • Research managers – Managers will be introduced to a technical and application domain ripe for revolutionary advances. 

Format

The Symposium will consist of three types of activities: introductory tutorials to provide participants with a common base of knowledge from which to work, approximately 15 short papers presenting current technical work, and two working sessions. Specifically: 
  • Tutorial: A short tutorial will be held during the first half of the second day. It will present the conceptual background of Soft Computing Techniques relevant for equipment service, focusing on unique aspects of the task. It will demonstrate how different AI techniques come together to tackle specific problems where one technique alone would have a hard time. Several examples will be shown. 
  • Short papers: Individual participants will present papers describing their current technical work. The organizing committee will attempt to divide papers evenly between purely technical papers and papers describing applications that have stretched the limits of current technologies. Each paper will be brief (about 15 minutes) with 10-15 minutes for discussion and questions after each paper. 
  • Working sessions: Interspersed with the papers, the symposium organizers will moderate two sets of working sessions. The topics will be finalized during the morning sessions to reflect the perceived interest of the audience. Proposed topics inlcude Modeling Issues, Diagnostic Information Fusion, Adaptation/ Reconfigurability and other write-in topics. The second set of working sessions on Tuesday has a similar policy. Proposed topics are Internet Supported Diagnosis, Diagnostic Ease of Authoring / Ease of Use, Autonomous Systems, Maintenance Planning and other write-in topics. In either case, the results will be presented after those work-outs. Each working session will run for 60 minutes. 
 
Preliminary Schedule 
Time Name Title Name Affiliation
Monday 
9:00-9:10
Opening Comments P. Bonissone 
K. Goebel
GE Corporate Research & Development
Monday 
9:10-9:45
Presentations & Discussion Integrated Diagnosis for Future Mobile Systems Baeker, Bernard & Forchert, Thomas DaimlerChrysler AG
Monday 
9:45-10:20
Concurrent Engineering Perspective of Maintenance Aspects through an Expert System Prototype Silva, Jonny da Federal University of Santa Catarina, Brazil
Monday 
10:20-11:00
Break
Monday 
11:00-11:25
Presentations & Discussion Distributed Equipment Monitoring with Multiply Sectioned Bayesian Networks Xiang, Y. & Geng, H. University of Regina, Sasketchwan
Monday 
11:25-11:50
RISO: An Implementation of Distributed Belief Networks Dodier, Robert University of Colorado at Boulder
Monday 
11:50-12:15
Acquisition and Transformation of Likelihoods to Conditional Probabilities for Bayesian Networks Skaanning, Claus & Jensen, Finn & Kjaerulff, Uffe & Madsen, Anders Hewlett Packard, Allborg University, Denmark
Monday 
12:15-2:00
Lunch
Monday 
2:00-2:30
Presentations & Discussion Condition-Based Monitoring of Motor-Pump Systems Using Model-Based Reasoning Chen, Yi-Liang & Provan, Gregory Rockwell Science Center
Monday 
2:30-3:00
Fault Propagation using Fuzzy Cognitive Maps Kipersztok, Oscar Boeing
Monday 
3:00-3:30
Towards Fuzzy Belief Nets Goebel, Kai & Agogino, Alice GE CR&D, UC Berkeley
Monday 
3:25-4:00
Break
Monday 
4:00-5:30
Working Session >> Modeling Issues 
>> Prognostics
all
Monday 
6:00-7:00
Reception
Tuesday 
9:00-10:30
Tutorial Soft Computing Techniques for Diagnostics and Prognostics Bonissone, Piero & Goebel, Kai GE Corporate Research & Development
Tuesday 
10:30-11:00
Break
Tuesday 
11:00-11:30
Presentations & Discussion Case-Based Troubleshooting Knowledge Management Gupta, Kalyan CaseBank Technologies, Inc.
Tuesday 
11:30-12:00
The Integrated Diagnostic System (IDS): Remote Monitoring and Decision Support for Commercial Aircraft - Putting Theory into Practice Halasz, Michael & Dube & Francois, Orchard, Bob & Ferland, R. National Research Council of Canada/Air Canada
Tuesday 
12:00-12:30
A Web-Based Intelligent Hybrid System for Fault Diagnosis Jha, G. & Hui, S., & Foo, S. Nanyang Technological University, Singapore
Tuesday 
12:30-1:00
Industrial Diagnosis by Hyperspace Data Mining Chen, N. & Zhu, Dongping & Wang, W. Zaptron Systems, Inc.& Salomon Smith Barney
Tuesday 
1:00-2:00
Lunch
Tuesday 
2:00-2:30
Presentations & Discussion From Free-Text Repair Action Messages to Automated Case Generation Farley, Benoit National Research Council of Canada/Air Canada
Tuesday 
2:30-3:00
Practical Optimization Considerations For Diagnostic Knowledge Representation Cornwell, Patricia & Suermondt, J. & Forman, G. & Kirshenbaum, E. & Seetharaman, A. Hewlett Packard
Tuesday 
3:00-3:30
Intelligent Knowledge Extraction from Service Notes for Equipment Diagnostics Varma, Anil & Ramani, Vipin GE CR&D
Tuesday 
3:30-4:00
Break
Tuesday 
4:00-5:30
Working Groups >> Text based Diagnosis
>> Internet Supported Diagnosis 
all
Tuesday 
6:00-7:30
Public Forum Summary of results fom all symposia
Wednesday 
9:00-9:30
Presentations & Discussion Fault Prognosis Using Wavelet Neural Networks Wang, Peng & Vachtsevanos, George Georgia Institute of Technology
Wednesday 
9:30-10:00
Warranty and Maintenance Decision Making for Gas Turbines Chao, Susan, Lee, Zu-Hsu & Agogino, Alice University of California at Berkeley
Wednesday 
10:00-10:30
Break
Wednesday 
10:30-11:30
Panel Discussion Where Do We Go From Here? Agogino, Alice & Bonissone, Piero & Goebel, Kai & Vachtsevanos, George
Wednesday 
11:30-12:00
Closing Comments Bonissone, Piero & Goebel, Kai
 

Proceedings 
The Proceedings can be obtained through AAAI Press. Please refer to  "AI in Equipment Maintenance Service and Support", 1999 AAAI Spring Symposium, Technical Report SS-99-04, AAAI Press, Menlo Park, CA, 1999. 


Abstracts must have been submitted by October 30, 1998 to one of the organizers. The length of an abstract should not exceed 200 words. Upon acceptance, submit to the symposium chair by January 29: 

  • the camera-ready paper (8 pages max.). Details on formatting suggestions can be found here.
  • the permission to distribute form.
  • audio/vidual requests other than an overhead projector.
Please use the following address: 
Kai Goebel
Symposium Chair, AAAI Spring Symposium
GE Corporate Research & Development
Information Technology Lab
K1-5C4A
One Research Circle
Niskayuna, NY 12309
USA
Submit to AAAI by January 29:  Here is information straight from AAAI: 

Electronic Submission Instructions 

 AAAI would like to maintain an electronic copy of your abstract and an electronic copy of your paper as well. 

To submit your abstract, please use the form on the AAAI web site or, to send the information by electronic mail, please follow these instructions: On the subject line, put the last name of the principal author, the name of the conference or event, the year, and the word "Abstract." For example: "Doe-SSS 99 Abstract". 

In the body of the message, put the title of your paper. Type this in initial caps, lower case. Don't type the title in all lower or upper case. See the example below. 

Skip a line, then put all the author's names in the order in which they appear in the paper. Put first name, then last name for each author. 

Skip a line, then put the word "Abstract" followed by the actual text of the abstract. 

Do not format your electronic abstract, and do not use LaTex, Tex, html, or any other coding method. 

Here's an example of a completed abstract: 

       Subject: Doe-SSS 99 Abstract  
       Investigation of the XYZ Algorithm  
       Jane Doe and John Doe  
       Abstract: The XYZ Algorithm is presented here.... 
Send your message to sss99-papers@aaai.org. 

Paper Submission 

 AAAI is now making a permanent electronic archive of all its publications. To aid us in this effort, you are are required to send us a PDF or PostScript version of your paper at the same time that you send in your camera-ready copy. Please follow these two steps to complete this process: 

1. Sign the permission to distribute form (also faxed upon request). By signing this form, authors give AAAI permission to distribute their paper. Authors do not relinquish copyright to their paper, nor do they need to request permission from AAAI before publishing their paper. The form does, however, give AAAI permission to distribute the paper in both hard copy and electronic format. (Please note that AAAI cannot distribute previously published material.) 

2. Your paper should be sent to sss99-papers@aaai.org. The subject line of your mail message, in all cases, should contain the last name of the principal author of the paper, followed by the conference name, the year, and the file format. For example: Doe-SSS99.pdf. Your paper can be sent in either of these two formats: 

a) A PDF file of your paper using Acrobat Distiller 3.0 or later. Distiller settings should be set to produce a high-end print PDF file; thus, automatic compression should be set to ZIP/JPEG low compression, the default resolution should be 1,240 dpi, compatibility should be set to Acrobat 3.0, and downsampling should be turned off. All fonts should be embedded, and the default page size must be set to letter (8.5 x 11 inches), not A4. Suggested settings can be found on AAAI's distiller settings web site

  

or 

b) If you do not have access to Acrobat Distiller, you may submit a PostScript file to us instead. Please set your PostScript file up using letter size (8.5 x 11 inches) as the default page size--not A4. If you are using a wordprocessor on a Windows or Macintosh platform, please embed all PostScript fonts except the standard families of Helvetica, Times, Symbol, and Palatino into your file. (Do not embed Truetype fonts.) If you use fonts other than Times and Helvetica, you lower the chance of successful conversion of your file to PDF; consequently, we strongly recommend that you use the Times Roman family of fonts. 

If you are unsure about how to create a PostScript file, please consult these instructions. or request help from someone at your site. 

About AAAI 

Founded in 1979, the American Association for Artificial Intelligence (AAAI) is a nonprofit scientific society devoted to advancing the scientific understanding of the mechanisms underlying thought and intelligent behavior and their embodiment in machines. AAAI also aims to increase public understanding of artificial intelligence, improve the teaching and training of AI practitioners, and provide guidance for research planners and funders concerning the importance and potential of current AI developments and future directions. Major AAAI activities include organizing and sponsoring conferences, symposia and workshops, publishing a quarterly magazine for all members, publishing a series of books, proceedings, and reports, and awarding grants and scholarships.

Version 1.1
Updated 3/24/99
Dates, times, and placesWhy a symposium now?Who should attend?FormatPreliminary ScheduleWorking SessionProceedingsAbout AAAI