Ahmad Ali Khasawneh

Pavement friction is a critical factor in ensuring road safety and performance. This review explores the correlation between asphalt pavement surface features and locked-wheel skid trailer friction values using Machine Learning (ML) algorithms. The study focuses on how ML algorithms, particularly Artificial Neural Networks (ANNs), can be utilized to analyze and predict pavement surface characteristics and their impact on skid resistance. The review provides an overview of common pavement surface features that influence skid resistance, such as texture and roughness, and describes the measurement techniques used to quantify these features. It also discusses the application of locked-wheel skid trailer testing as a method for measuring pavement friction and highlights the potential of ML algorithms in analyzing skid resistance data. The review concludes with a discussion on the challenges and future directions of using ML algorithms and ANNs in pavement engineering studies, emphasizing the importance of study in improving road safety and pavement performance.

Asphalt concrete pavement is considered a versatile and affordable material for constructing durable roads. However, despite its obvious advantages, this type of infrastructure is highly susceptible to various types of distress, which can affect its performance and service life. Moisture-induced damage, sometimes known as "stripping," has arisen as a major concern among these distress mechanisms in recent years. Moisture-induced damage occurs when water infiltrates the pavement, weakening the asphalt-aggregate bond and causing a variety of distresses that compromise the performance and durability of the pavement. Thus, understanding the mechanisms that cause moisture-induced damage is crucial for developing effective preventive and mitigation strategies. This paper aims to provide a comprehensive review of the moisture-induced damage phenomenon, focusing on the stripping mechanism in asphalt concrete pavements. Compared with previous publications, this article highlighted the main predictive models for predicting moisture-induced damage and served as a valuable resource for researchers, engineers, and policymakers involved in the design, construction, and maintenance of asphalt concrete pavements, fostering continued advancements in the field, and promoting sustainable and resilient infrastructure systems. Finally, it is recommended to use machine learning (ML) and artificial neural networks (ANN) for moisture-induced damage prediction.