Home | Research | Research Programs | Artificial Intelligence Technologies | Integrated Diagnostic System (IDS)

Artificial Intelligence Technologies

Integrated Diagnostic System (IDS)

Image representing the IDS project - Click for full view

The recently completed Integrated Diagnostic System (IDS) project was undertaken to research, develop, integrate and test a number of advanced software-based diagnostic techniques for the maintenance of complex equipment.

The resulting IDS prototype includes the ability to categorize real-time aircraft-generated messages in order to flag potential problems, thus allowing maintenance technicians to respond more quickly and efficiently.

Ultimately, the IDS tool could be applied in many other technical domains beyond the aircraft industry, including:

• power generation
• manufacturing
• aerospace
• forestry
• transportation

The IDS prototype is the result of research efforts focused on exploring the pragmatic issues surrounding the construction of useful, maintainable knowledge based systems. In the early 90s — when research on IDS began — researchers were preoccupied with two questions:

1. how can representation and reasoning techniques be selected?
2. how can knowledge based systems be effectively integrated into their environments (i.e., match the various users’ requirements, exploit existing IT infrastructure)?

With the success of IDS and other related projects, researchers have now shifted their focus to the problems of streamlining/supporting the acquisition and management of the knowledge bases that drive the decision support/automation processes.

The IDS prototype, along with its infrastructure and data resources, continues to provide an excellent platform upon which to base further research. Some important issues currently under investigation include:

• knowledge extraction from manuals (KDS) — i.e., exploring more refined ways in which to extract information from free form and structured text
  
  
• automatic case creation and maintenance (CBR)
  
  
• knowledge and data driven constructive induction and automatic trend recognition (ADAM) — i.e., experiments with automatic feature extraction from aircraft performance data to provide richer symptom sets for case retrieval.

The development of IDS was carried out in collaboration with Air Canada and has been commercialized by Mxi Technologies Ltd.

Related NRC-IIT Publications

Research Contact

Robert Orchard (Bob)
Group Leader
Integrated Reasoning
NRC Institute for Information Technology
1200 Montreal Road
Building M-50, Room 366
Ottawa, ON K1A 0R6
Telephone: +1 (613) 993-8557
Fax: +1 (613) 952-0215
E-mail: Robert.Orchard@nrc-cnrc.gc.ca

Business Contact

Dr. George Forester
Business Development Officer
Business Development Office, NCR
NRC Institute for Information Technology
1200 Montreal Road
Building M-50, Room 203
Ottawa, ON K1A 0R6
Telephone: +1 (613) 993-3478
Fax: +1 (613) 952-0074
E-mail: George.Forester@nrc-cnrc.gc.ca