Aviation reports indicate that between 1988 and 2019 there were 292 human deaths and 327 injuries that had been reported from wildlife strikes with airplanes. To minimize these numbers, a new approach to airport Wildlife Hazard Management (WHM) is presented in the following article. The proposed solution is based on the data fusion of thermal and vision streams, which are used to improve the reliability and adaptability of the real-time WHM system. The system is designed to operate under all environmental conditions and provides advance information on the fauna presence on the airport runway. The proposed sensor fusion approach was designed and developed using user-driven design methodology. Moreover, the developed system has been validated in real-case scenarios and previously installed at an airport. Performed tests proved detection capabilities during day and night of dog-sized animals up to 300 meters. Moreover, by using machine learning algorithms during daylight, the system was able to classify person-sized objects with over 90 % efficiency up to 300 meters and dog-sized objects up to 200 meters. The general accuracy of the threat level based on the three safety zones was 94 %. © 2022 Kauno Technologijos Universitetas. All rights reserved.