Abstarct: Roads are one of the most important assets of any country and a large part of the country's budget is spent annually on repair and maintenance operations. A large part of the cost of road repair and maintenance operations is spent on repairing cracks. The self-healing potential of asphalt is one of the factors that can increase the useful life of asphalt pavement. When the pavement is at a higher temperature, the repair increases, so the repair of the asphalt mixture itself is temperature dependent. Microwave heating and induction heating are used to increase self-healing capacity by increasing the temperature. Also, to increase heating efficiency and improve self-healing, different types of mextal wastes (steel fiber, mextal shaving, steel slag, etc.) are added to the asphalt mixture. Due to the fact that laboratory activities are time consuming and costly, the use of artificial neural networks in predicting laboratory results in recent decades has been considered by researchers. In pavement engineering, artificial neural networks have been used to predict the fatigue life of asphalt mixtures, to predict the nonlinear modulus of pavement laxyers, etc. The purpose of this study is to consider the factors affecting the self-healing index (such as type of additive, percentage of additive, gradation of aggregate, type of bitumen, crack repair cycle, etc.) as inputs of artificial neural network in MATLAB software and self-healing index as output, to achieve a suitable neural network and use this network to predict self-healing index. Multilxayer perceptron neural network (MLP), multilxayer perceptron neural network with particle swarm optimized algorithm (PSO), radial basis function neural network (RBF) and statistical analysis with SPSS software have been used in this study and the results of these methods together were compared. The results showed that multilxayer perceptron neural network (MLP) has better performance in predicting self-healing index than other methods.