To understand traffic phenomena and further to solve congestion problems, various traffic flow models have been developed. Recently, there have been many attempts to manage congestion at the network level using macroscopic fundamental diagrams (MFD), because these have the advantage of enabling effective management of the network traffic while reducing the complexity of the model. In this study, the author proposes a method to utilize large-scale data collected at the micro level of individual vehicles for a macroscopic traffic management strategy at the city level. First, this study investigates how to process and modify the transportation big data, and use it to estimate the empirical MFDs for a real metropolitan city. This study analyzes the bifurcation phenomenon in empirically estimated MFDs to find causal factors, and create a method to improve network performance based on the findings. To this end, this study proposes a utility-maximized route guidance strategy considering individual route choice preferences. Applying the proposed strategy through Model Predictive Control (MPC)-based simulation yielded positive effects on both network performance and drivers’ utility. Accordingly, it can be expected that the proposed strategy will increase drivers’ route compliance rate. The simulation also showed that the higher the compliance rate, the better the effectiveness of the proposed strategy.