Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1 SUMMARY 1h January 1993, Me University of Idaho, in cooperation with Kittelson and Associates, Ruhr University, and Queensland University of Technology, initiated work on NCHRP 3~6 to develop new capacity and level of service analysis procedures for unsignalized Intersections. This report, the first of two volumes describing We work completed for this project, documents the work completed for two-way stop-controDed (TWSC) Intersections. Volume two includes eight chapters covering the work completed for as-way stop-controlled (AWSC) Intersections. A new data base descnb~ng traffic operations at TWSC intersections was assembled as part ofthis study. A total of 68 unique sites wer~videotapeddurLng 79 different time periods. In most cases, four video cameras were used at each site producing a total of over 300 individual videotapes. A total of 128 hours of usable data was acquired. The data base includes information from all geographic sectors of We United States and covers a wide range of traffic operational and geometric conditions. Two types of models were used to analyze traffic operations at TWSC intersections: gap acceptance models and empirical models. The theoretical background of venous capacity models, delay models, and cntical gap estimation procedures were documented Candidate models and procedures were selected based on certain criteria and then tested against field data. Critical gap and foDow-up time are the two major parameters used by various gap acceptance models. The maximum likelihood procedure was recommended for estimating critical gaps. Follo~up time was estimated directly from field data. A completely new set of cntical gaps and foDow-up bmes were recommended, based on the new data set collected for U.S. conditions. Gap acceptance capacity models developed by Harders (used In the 1985 HCM) and Siegloch (used In the 1994 HCM) were tested. Harders' model, whose results were marginally better Han the Smooch model, is recommended for future applications. A comprehensive calibration and validation process showed Hat the basic computational procedures used in the 1994 HCM are valid; these procedures are recommended to evaluate TWSC intersections. However, the following changes are recommended as additions to He basic procedure: . . new values for critical gaps and follow-up times, a weighting factor of 2 to be applied to major street left turn flows to calculate conflicting volumes, a procedure to account for the effect of upstream signals, a mode! to account for the effect of a two-stage gap acceptance process, flared approach, and pedestrians, and delay calculations for the major sheet through vehicles. A t~me-dependent delay mode! developed by Troutbeck and Akcelik is recommended. (This is He same mode! used in the 1994 HCM). The mode! provides estimated delay in queue at TWSC intersections. Other topics included in this document are discussions of simulation models, MUTED signal warrants, and LOS cnteria. Testing ofthe available simulation models showed that only KNOSIMO, developed by Briton, gave satisfactory results. Selection of intersection control type based on the average intersection has demonstrated consistency with the peak hour signal warrants given by MUTCD provided that burning movement proportions are "average". Finally, while delay can be a good indicator to define intersection LOS at TWSC intersections, it is probably more reasonable to define LOS with consideration given to both delay and queue length.