An arterial bandwidth maximization model with phase release pattern optimization under symmetrical phasing

Bandwidth-based traffic signal coordination has long been regarded as an efficient and economical strategy to mitigate traffic congestion. However, under symmetrical phasing, existing arterial bandwidth maximization methods mainly focus on optimizing the common cycle length and offsets, which results in a limited optimization space, making it difficult to simultaneously yield satisfactory bandwidths in both directions of the arterial. To address this limitation, this paper proposes the PRPO-Band model, an arterial bandwidth maximization model with phase release pattern optimization under symmetrical phasing. The phase release pattern is defined as the number of releases of coordinated phase and the green time allocated to each release, which are optimized jointly with the common cycle length and offsets. The proposed PRPO-Band model is formulated as a mixed-integer nonlinear program (MINLP). Through linearization, the PRPO-Band model is transformed into a mixed-integer quadratically constrained program (MIQCP), which is solved using the Gurobi optimizer. The bandwidth results show that the proposed PRPO-Band model can generate a larger bandwidth at the arterial level than the two baseline models. Sensitivity analysis shows that the large positive number M has virtually no influence on the optimal value of the proposed model; however, the optimal value decreases when either the lower limit on the green time of the coordinated phase or the lower limit on bandwidth exceeds a certain threshold, while it remains unchanged under certain combinations of these two parameters.