Comparative Analysis of Wear Models for Accurate Wear Predictions

Abstract

The development of innovative technologies and the employment of diverse material compositions have contributed to the enhancement of wear prediction methods. However, the accurate forecasting of service life and the identification of critical influencing factors remain challenging due to the complex interactions governing wear behaviour. Throughout history, various methodological approaches have been developed to model wear, primarily categorised into analytical calculations and experimental investigations. Analytical methods, including Archard’s equation and its variations, provide a theoretical basis for wear estimation. However, these models frequently depend on empirical coefficients derived from extensive experimentation, which restricts their predictive accuracy. Moreover, classical wear models do not fully account for material fatigue effects and 3D surface texture parameters, which are critical for solving complex engineering problems. Recent advancements have sought to address these limitations by integrating probabilistic surface modelling, fatigue-based degradation theories, and numerical simulations to enhance wear predictions. Experimental investigations remain essential for validating analytical models, as they provide empirical data necessary for parameter calibration. However, these experiments require specialised equipment and are often time-consuming and costly. The integration of modern measurement tools and numerical simulations, such as finite element analysis (FEA) and machine learning-based models, presents a promising direction for improving wear predictions. This review highlights the strengths and limitations of existing wear models and emphasises the need for further refinement of analytical approaches to incorporate fatigue wear mechanisms, real surface roughness effects, and environmental influences for more accurate and reliable wear assessments.