Pavement management systems play a vital role in the development of a country as it is a very important part of the economy. Maintaining a good quality of the road is the key duty of the road authorities. They require methods for pavement-related data collection and analysis to evaluate its condition. Machine learning (ML) methods can be utilized for defect classification from an image, defect recognition and segmentation in the assessment of pavement distress. This paper presents an overview of the machine learning techniques used to analyze pavement condition data. Moreover, information collection methods and pavement condition indices are also studied from the point of view of ML algorithms. Future research directions are also presented by highlighting the limitations of using ML techniques for the assessment of pavement conditions. © 2022 IEEE.