Road networks are increasingly being monitored to understand and improve travel time reliability. However, traditional calculation methods have disregarded the interdependent nature of links in a network by focusing on individual link unreliability. This project proposes to overcome those concerns by (1) exploring a systemic approach to calculating network travel time reliability, (2) screening and ranking links that are highly correlated with network unreliability, and (3) clustering such links for effective selection and implementation of mitigation measures, and for improved before after studies. The systemic approach also preserves the timestamps associated with unreliable network travel times, and helps identify their causes.