Enhanced traffic delay models for isolated intersections and congested arterials

Traffic delay models play significant roles in real-time intelligent transportation systems, traditional transportation planning systems, and the field of highway capacity analysis. Different types of roadways and control devices have associated with them different delay functions. In this study, isolated signalized intersections with vehicle-actuated control and congested signalized arterials with pretimed control are identified as two important areas requiring improved delay estimation methods. Enhanced delay models that provide a more accurate estimation of signal performance are presented.; A macroscopic, stochastic, cycle-by-cycle simulation model developed in earlier work for pretimed control operation was modified by the author to study the capacity and delays for the basic fully-actuated operation. The overflow delay was estimated as the difference between the simulated approach delay and the computed Webster's uniform delay based on simulated signal timing parameters. Separate models are calibrated for different values of controller unit extensions. The analytical models for estimating average signal timing parameters for basic fully-actuated control operation are also derived.; In the case of signalized arterials, signal approaches release traffic at no more than some maximum value corresponding to the signal capacity in congested flow conditions. This is referred to as signal metering effect. This effect reduces demand at the downstream intersection and results in more uniform arrivals from cycle to cycle. Thus, isolated intersection delay formulae based on the assumption of random arrivals may seriously overestimate delays for signalized arterials. In this study, analytical overflow delay models for signalized arterials accounting for signal metering effect are developed. The analytical models are founded on the random queue model and it could be considered as a quantitative measure of the upstream metering effect on delay in congested signalized arterials.; The format of the delay models developed in this work is consistent with the format of the 1985 HCM delay models. The enhanced models, when effectively validated by field data, can be substituted for the current HCM models, which have many limitations. The enhanced models can also provide improved guidelines for transportation planners and engineers.