We systematically review the recent studies on wear and friction of pipeline walls caused by solid particle impact using AI related techniques. We examine the existing theoretical and experimental approaches, and analyze the advantages of Artificial Intelligence methods and their complementarity with classic research methods. According to the classification of data sources, the prediction results of Artificial Neural Networks (ANNs) based on experimental data or CFD simulation are discussed separately. The latest cases of erosion research using Support Vector Machine, Random Forest and other methods are also briefly introduced. Recent research conclusions show that AI method has high potential for application in the prevention of pipeline erosion, and will play an increasingly important role in the improvement of pipeline integrity management and the acceleration of
Intelligent Pipeline Construction in the future.