Simulation-based estimates of delays at freeway work zones

Work zone related traffic delay is an important cost component on freeways with maintenance activities. This study demonstrates that delays may be underestimated by using the deterministic queuing theory. Computer simulation is a valuable approach of estimating delay under a variety of existing and future conditions. However, a single simulation run, which can be quite costly in terms of both computer and analyst time, produces a delay estimate for only one traffic level under one set of conditions. A method is developed in this paper to approximate delays by integrating limited simulation data, obtained from CORSIM and the concept of deterministic queuing theory, while various geometric conditions and time-varying traffic distribution are considered. A calibrated and validated simulation model that can reflect work zone traffic operations on a segment of Interstate I-80 in New Jersey is used to generate data for developing the proposed model. The comparison of delays estimated by the deterministic queuing model and the proposed model is conducted, while factors affecting the accuracy of the delay estimates are discussed.