The trucking industry is an irreplaceable sector of our economy. Over 80% of the world population relies on it for the transportation of commercial and consumer goods. In the US alone, this industry is responsible for over 38% of fuel consumption as it distributes over 70% of our freight tonnage. In the design of these vehicles, particular emphasis has been placed on equipping them with a strong engine, a relatively comfortable cabin, a spacious trailer, and a flat back to improve loading efficiency. The geometrical design of these vehicles makes them prone to flow separation and at highway speeds overcoming aerodynamic drag accounts for over 65% of their energy consumption. The flat back on the trailer causes flow to separate, which generates a turbulent wake. This region is responsible for a significant portion of the aerodynamic drag and currently the most popular solution is the introduction of flat plates attached to the back of the trailer to push the wake downstream. These passive devices improve the aerodynamic performance of the vehicle, but leave opportunities for significant improvement that can only be achieved with active systems. The current procedure to analyze the flow past heavy vehicles and design add-on drag reduction devices focuses on the use of wind tunnels and full-scale tests. This approach is very time consuming and incredibly expensive, as it requires the manufacturing of multiple models and the use of highly specialized facilities. This Dissertation presents a computational approach to designing Active Flow Control (AFC) systems to reduce drag and energy consumption for the trucking industry. First, the numerical tools were selected by studying the capabilities of various numerical schemes and turbulence model combinations using canonical bluff bodies. After various numerical studies and comparisons with experimental results, the Jameson-Schmidt-Turkel (JST) scheme in combination with the Shear-Stress-Transport (SST) turbulence model were chosen. This combination of tools was used to study the effect of AFC in the Ground Transportation System (GTS) model, which is a simplified representation of a tractor-trailer introduced by the US Department of Energy to study the separation behind this type of vehicle and the drag it induces. Using the top-view of the GTS model as a two-dimensional representation of a heavy vehicle, the effect that the Coanda jet-based AFC system has on the wake and integrated forces have been studied. These two-dimensional studies drove the development of the design methodology presented, and produced the starting condition for the three-dimensional Coanda surface geometry and the jet velocity profile. In addition, the influence in wake stability that this system demonstrated when operating near its optimum drag configuration, allowed for the decoupling of time from the three-dimensional design process. A design methodology that minimizes the number of required function evaluations was developed by leveraging insights obtained from previous studies; using the physical changes in the flow induced by the AFC system to eliminate the need for time integration during the design process; and leveraging surrogate model optimization techniques . This approach significantly reduces the computational cost during the design of AFC drag reduction systems and has led to the design of a system that reduces drag by over 19% and power by over 16%. In the US trucking fleet alone, these energy savings constitute 8.6 billion gallons of fuel that will not be burned and over 75 million tons of CO2 that will not be released into the atmosphere each year.