ABSTRACT A multi-layer inductive healing layer (ML-IHL) based on the magnetic field distribution characteristics was designed to solve gradient healing behaviors of conventional single-layer inductive healing layer (CSL-IHL). Two representative conductive metals, including waste steel shaving and steel wool fibers, were used to prepare two kinds of ML-IHL, which is a 2-layer system composed of asphalt mixtures with different conductive metal contents. Then infrared images of the temperature distribution of ML-IHLs and CSL-IHL were recorded to analyze the temperature changes at different depths. Subsequently, the temperature coefficient of variation and radar chart method was used to evaluate the temperature gradient. Results show that as the difference in the conductive metal contents between the two lays of the ML-IHL increases, the temperature gradient gradually decreases. Also, it is recommended that the optimum content in the upper layer containing waste steel shavings or steel wool fibers is 4% or 2%, and the optimum content in the lower layer is 8%. Finally, the inductive healing test of the optimal scheme shows that the healing effect of the upper and lower layers of both ML-IHL has been significantly improved, and the healing rate difference between the two layers is all less than 2%.