Two research efforts devoted to the potential use of computers in the control of demand-responsive transportation (DRT) systems are reported: a computer simulation model to test alternative computer control algorithms and to predict system performance; and a set of control procedures in which (a) the immediate assignment of each request was made to the current 'tour' of the best vehicle, (b) the assignment was based on feasibility conditions, under which each user receives service within specified bounds, and (c) the determination of the best assignment was based on the minimization of total service times for current and future passengers. These control procedures were tested by a simulation model and were found to perform well in intuitive grounds and relative to other proposed algorithms. However, since no optimal solution algorithm has been developed, absolute statements about performance were not possible. A demonstration project of the concept was mounted in Haddonfield, New Jersey, to obtain a market test of the service concept and to obtain data on the potential of computer dispatching. The assumptions and simplifications of the real-world system required in the design of the simulation model are discussed, and the findings on modeling assumptions are described. Areas in which improved performance might be achieved are: inflexibility of hand constraints, objective function as a true reflection of utility, handling of advanced and periodic requests, constraint of vehicle position at future time, restriction of certain vehicles to given zones, preassignment capability, scheduling at start and end of driver and vehicle shift, and gearing of algorithm to underused system. The study shows that the simulation model can accurately predict system performance providing that vehicle in-service times are used. The lower demand densities at Haddonfield suggest that the economies of scale possible with these systems cannot yet be realized - and that productivities of 5 to 8 passenger trips per vehicle hour are more realistic than previously cited ranges of 9 to 13. Current research suggests that it is both feasible and desirable for the computer algorithms to achieve better service and to allow the operation of large integrated DRT and conventional transit systems.

  • Record URL:
  • Availability:
  • Supplemental Notes:
    • Presented at the Fifth Annual International Conference on Demand-Responsive Transportation Systems conducted by the TRB, Nov. 11-13, 1974, Oakland, Calif.; and co-sponsored by American Public Transit Association, California DOT, Alameda-Contra Costa Transit, MIT, UMTA and Technology Sharing Program of U.S. DOT. Distribution, posting, or copying of this PDF is strictly prohibited without written permission of the Transportation Research Board of the National Academy of Sciences. Unless otherwise indicated, all materials in this PDF are copyrighted by the National Academy of Sciences. Copyright © National Academy of Sciences. All rights reserved.
  • Corporate Authors:

    Transportation Research Board

    500 Fifth Street, NW
    Washington, DC  United States  20001
  • Authors:
    • Wilson, Nigel H M
  • Conference:
  • Publication Date: 1975

Media Info

  • Media Type: Digital/other
  • Features: Tables;
  • Pagination: pp 104-109
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00126181
  • Record Type: Publication
  • Files: TRIS, TRB, ATRI
  • Created Date: Dec 3 1981 12:00AM