The objective of this research is to develop a general methodological framework for planning and evaluating the effectiveness of highway reconstruction strategies on the system's performance measures, in particular safety, mobility, and the total cost of network rehabilitation. Transportation networks are characterized by uncertainty that stems from different sources and transportation planners should consider risks involved in uncertainty in model parameters. In this research, Conditional Value at Risk (CVaR) is used to quantify and measure the risk in pavement performance and travel demand. First, a method of constructing risk based rehabilitation policies for a network of pavement facilities that ensures a specific quality level is introduced. Second, to enhance network mobility, several optimization models to minimize travel time for all paths connecting the same origin and destination pair is presented. Finally, to improve network safety during capacity expansion decisions, two models are presented to reduce accident rate by changing the ratio of flow to the link capacity to reduce injury and fatal accidents and property damage accidents.