Optimum Cycle Length Model for Fixed-Time Signalized Intersections with Meta-Heuristic Methods

Recently, growth in population and economy has increased the use of vehicles on highways. Depending on this, the capacity of intersections becomes increasingly insufficient. Due to inefficient operation of intersections, the delay, fuel consumption, and emission release increase, and also the driver behaviors are negatively affected. It is possible to minimize these problems by correctly determination of optimum cycle length and adjustment of signal times. In this study, optimum cycle length models have been developed using Artificial Bee Colony Algorithm (ABC). In addition, Differential Evolution Algorithm (DE) has been used to determine the cycle length which has the minimum delay. In addition to the Calibrated Webster model, exponential and quadratic forms have been developed. All developed models have statistically better performance than the Webster model and also exponential model has illustrated the best performance. It has been seen that these models can be used as an alternative estimating cycle length model by overcoming the deficiency of Webster model which is insufficient especially in high traffic volumes