KNOWLEDGE-BASED SYSTEMS FOR TRANSPORT NETWORK ANALYSIS: A FIFTH GENERATION PERSPECTIVE ON TRANSPORT NETWORK PROBLEMS

This paper considers the formative steps in the development of an expert system for route selection in transport networks. It discusses the application of knowlegde-based system (KBS) technology to the production of an expert system to provide route guidance information for some groups of urban travellers. There is a focus on the possibilities for applying a "fifth generation" programming language, such as PROLOG, to the solution of network path selection problems, and the maintenance of a dynamic network database. There are significant arguments both for and against the use of the fifth generation languages, and the alternatives are discussed in speculative terms, on the basis of trials and experiments with PROLOG. The paper develops the specifications for a route guidance KBS for urban road travel, noting that one particular concern is that of providing the most suitable advice to a given traveller, rather than the system-wide rating of "best" advice. Driver attitudes, behaviour, preferences, and vehicle characteristics differ widely, and the degree of satisfaction with any advice provided would depend largely on how closely the expert system could match the preferences of the individual traveller. Thus the system would need to gauge the characteristics of the user as well as those of the network under consideration. Further, prototype route guidance KBS are probably better directed at some clearly defined groups of urban travellers, such as commercial vehicle operations, rather than as community-wide systems.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/09658564
• Corporate Authors:

  Pergamon Press, Incorporated

  Headington Hill Hall
  Oxford OX30BW,    
• Authors:
  • TAYLOR, MAP
• Publication Date: 1990-1

Media Info

• Features: Figures; References;
• Pagination: p. 3-14
• Serial:
  • Transportation Research Part A: Policy and Practice
  • Volume: 24A
  • Issue Number: 1
  • Publisher: Elsevier
  • ISSN: 0965-8564
  • Serial URL: http://www.sciencedirect.com/science/journal/09658564

Subject/Index Terms

• TRT Terms: Buses; Commercial vehicles; Computer programs; Expert systems; Route choice; Route guidance; Transportation planning; Urban transportation
• Subject Areas: Highways; Operations and Traffic Management; Planning and Forecasting; Public Transportation;

Filing Info

• Accession Number: 00492280
• Record Type: Publication
• Files: TRIS, ATRI
• Created Date: Mar 31 1991 12:00AM