SINGLE-COMMODITY AND MULTI-COMMODITY NETWORK IMPROVEMENT PROCEDURES

In this report a multi-modal freight transport improvement problem is formulated, and a heuristic solution procedure is developed for large scale problems. Arcs on a multi-modal network are modified to minimize the sum of investment and shipper disutility. The model includes a mode abstract multinomial logit model and convex arc transport characteristic improvement functions. The procedure is based on the general Continuous Optimal Adjustment heuristic suggested by Steenbrink; a major component is a concave transportation assignment problem, for which two methods are developed: a Dantzig-Wolfe decomposition to solve an arc-path form of the general multi-commodity fixed charge network flow problem, and a heuristic based on the local optimum seeking procedure of Yaged. The latter was programmed and thirty test runs were made on a problem derived from the Multi-State Transportation Corridor research program. Conclusions reached are: the solution procedure is viable; although times are long, this is not unusual for problems of this size or design construction projects of this scale; solution times may be shortened considerably by proper selection of parameters. As a final step, the procedure is extended to include multiple transport commodity classes.

  • Corporate Authors:

    University of Georgia, Experiment

    Georgia Experiment Station
    Experiment, GA  USA 

    Office of the Secretary of Transportation

    1200 New Jersey Avenue, SE
    Washington, DC  USA  20590
  • Authors:
    • Mullens, M A
    • SHARP, G P
  • Publication Date: 1979-3

Media Info

  • Pagination: 289 p.

Subject/Index Terms

Filing Info

  • Accession Number: 00197510
  • Record Type: Publication
  • Source Agency: National Technical Information Service
  • Report/Paper Numbers: DOT-OST-80050-1C Spec Rpt.
  • Contract Numbers: DOT-OST-80050
  • Files: NTIS, TRIS, USDOT
  • Created Date: Jul 31 1979 12:00AM