| The challenge of sustainable transportation development is to accomplish a balance between mobility and environmental sustainability. To help accomplish the balance, this research primarily addresses one of the most general unsustainable impacts of transportation -- congestion on urban streets.;Besides the optimization of basic signal timings (cycle, offset, splits and phase sequence), the optimization of advanced Transit Signal Priority (TSP) settings is enabled to additionally improve transit operations without worsening overall traffic conditions. The optimization of the settings for the two most common TSP strategies, green extension and early green, increases the reliability of transit service and attracts more customers, which reduces total number of vehicles on streets.;Another unsustainable impact of transportation is addressed with the research. To enable microscopically estimated instantaneous fuel consumption or vehicular emission to be used as an objective function during optimization process, VISGAOST is integrated with Comprehensive Modal Emission Model (CMEM), which provides estimates of second-by-second emissions of individual vehicles based on modes of a common driving cycle.;An additional measure which provides successful traffic signal optimization is achieved by integration of VISGAOST and VISSIM - ASC/3 Software-In-The-Loop-Simulation suite. The optimization of exact traffic signal timings from field in a risk-free environment eliminates extensive field timing tune-ups usually required after an optimization in the lab environment.;A traffic signal timing optimization tool, which finds the best solution through an evolution of initial signal timings while relying on the evaluations of the underlying high-fidelity microscopic simulator representing real-world traffic conditions, is developed. The tool is described as VISsim-based Genetic Algorithm Optimization of Signal Timings (VISGAOST), which identifies the evolution algorithms and the traffic microsimulator integrated within software.;The results of the experiments, verified with multiple randomly seeded simulations, show superiority of the optimized signal timings for the investigated performance measures. The statistically significant improvements in delays, number of stops, fuel consumption and CO2 emission range from 1% to 35%, mostly depend on the quality of initial signal timings and the objective function selected during individual signal timing optimization. |
