An Investigation of Convergence and Accuracy Properties of Latin Hypercube Sampling Technique for Traffic Equilibrium Problem under Capacity Uncertainty

Traffic Equilibrium Problem under Capacity Uncertainty (TEPCU) has drawn significant interest in recent years mainly due to the desire for incorporating stochastic nature of link capacities in the transportation planning process. The main driving force behind this new interest in incorporating uncertainties is the recognition of the need for more robust transportation plans especially for critical scenarios such as emergency management and evacuation. A common approach for solving TEPCU is to use a sampling technique by randomly selecting subsets of the uncertainty set to obtain approximate solutions. Latin Hypercube Sampling (LHS) is one of the most frequently used sampling methods which can provide accurate approximation. However, the main concern when using LHS is the required sample size which is found to be important in the application of LHS for TEPUC due to high computational times required for large networks. The main objective of this paper is to conduct an in depth analysis of ¡°convergence¡± and ¡°approximation accuracy¡± properties of LHS. Several computational tests are conducted using two different networks to determine the efficient sample size to obtain an accurate approximate solution of TEPUC at a given level of confidence. Efficiency of the sampling scheme, measured by the error of the estimate and approximation accuracy, is determined by comparing the result of LHS technique with the benchmark result obtained from the Monte Carlo (MC) method. The results provide us with a better understanding of the requirements of an appropriate experimental design for applying LHS to TEPUC on large transportation networks.


  • English

Media Info

  • Media Type: DVD
  • Features: Appendices; Figures; References; Tables;
  • Pagination: 15p
  • Monograph Title: TRB 89th Annual Meeting Compendium of Papers DVD

Subject/Index Terms

Filing Info

  • Accession Number: 01152551
  • Record Type: Publication
  • Report/Paper Numbers: 10-3504
  • Files: BTRIS, TRIS, TRB
  • Created Date: Jan 25 2010 11:46AM