The presence of work zones on freeways causes traffic congestion and creates hazardous conditions for commuters and construction workers. Traffic congestion resulting from work zones causes negative impacts on traffic mobility (delay), the environment (vehicle emissions), and safety when stopped or slowed vehicles become vulnerable to rear-end collisions. Addressing these concerns, a data-driven approach was utilized to develop methodologies to measure, predict, and characterize the impact work zones have on Michigan interstates. This study used probe vehicle data, collected from GPS devices in vehicles, as the primary source for mobility data. This data was used to fulfill three objectives: develop a systematic approach to characterize work zone mobility, predict the impact of future work zones, and develop a business intelligence support system to plan future work zones. Using probe vehicle data, a performance measurement framework was developed to characterize the spatiotemporal impact of work zones using various data visualization techniques. This framework also included summary statistics of mobility performance for each individual work zone. The result was a Work Zone Mobility Audit (WZMA) template which summarizes metrics into a two-page summary which can be utilized for further monitoring and diagnostics of the mobility impact. A machine learning framework was developed to learn from historical projects and predict the spatiotemporal impact of future work zones on mobility. This approach utilized Random Forest, XGBoost, and Artificial Neural Network classification algorithms to determine the traffic speed range for highway segments while having freeway lane-closures. This framework used a distribution of speed for each freeway segment, as a substitute for hourly traffic volume, and were able to predict speed ranges for future scenarios with up to 85% accuracy. The ANN model reached up to 88% accuracy predicting queueing condition (speed less than 20 mph), which could be utilized to enhance queue warning systems and improve the overall safety and mobility. Mobility data for more than 1,700 historical work zone projects in state of Michigan were assessed to provide a comprehensive overview of the overall impact and significant factors affecting the mobility. A Business Intelligence (BI) approach was utilized to analyze these work zones and present actionable information which helps work zone mobility executives make informed decisions while planning their future work zones. The Pareto principle was also utilized to identify significant projects which accounted for a majority of the overall impact. Chi-square Automatic Interaction Detector, CHAID, algorithm was also applied to discover the relationship between variables affecting the mobility. This statistical method built several decision-trees which could be utilized to determine best, worst, and expected consequence of different work zone strategies.