| In this dissertation, a linear mathematical model is developed to model the traffic flow across a freeway facility. This mathematical model is capable of addressing the optimization of any strategy that has some effects on the facility demand or capacity such as ramp metering. In this research, the mathematical model is configured and used to optimize ramp metering operations across a freeway facility, which is one of the Active Traffic Demand Management (ATDM) strategies that is utilized to optimize demand across a freeway facility to avoid congestion. However, with appropriate changes, the proposed model can optimize other facility demand or capacity strategies such as hard shoulder running. The model is a planning-level method for analyzing and determining the time varying metering rates appropriate to be used and implemented on freeway facilities. The stochastic nature of the freeway segments capacity has been incorporated in the proposed mathematical model, which leads to reliability evaluations. The linkage between time and space enables the proposed mathematical model to account for vehicle travel time on the facility that leads to adjust the start and end time of the metering operation. Feasibility studies on the proposed mathematical model lead to evaluation of applicability of ramp metering in the freeway facilities and estimation of freeway reliability. West bound Highway I-580 in California is used to implement the optimal ramp metering rates as a case study. The results of the proposed algorithm are compared to the results of ALINEA, a widely used ramp metering algorithm in the literature. The comparison of the two approaches indicates that the proposed method can perform at least as well as ALINEA. Thus, not only the proposed mathematical model gives optimal ramp metering rates compared to ALINEA, but it also can provide the option of adjusting the level of performance reliability of the freeway system. |
