This study proposes a fuel efficiency prediction model using a newly defined fuel efficiency index. The process is divided into three steps: (1) define a new fuel efficiency index after identifying the limitations of the existing fuel efficiency index; (2) set up a fuel efficiency prediction model; and (3) formulate possible actions for the airline to enhance fuel efficiency and reduce carbon emissions based on the model established in the previous step. The fuel efficiency prediction model was established using the actual flight data of Airbus 330-300 (engine type: PW4168A). The flight data were obtained from the fuel management and information system of an airline. The multiple regression model is used to identify the independent variables affecting the fuel efficiency and the degree of influence of each variable. The results indicate that variables such as payload, aircraft fuel mileage deterioration, center of gravity, extra fuel loaded, flight distance, and outside air temperature affect the fuel efficiency. Some variables can be controlled and managed by airlines, others are not. The proposed fuel efficiency prediction model is expected to be utilized as a measurable method for enhancing the fuel efficiency and reducing the carbon emissions.