Decoding Historical Stair Usage through Wear Analysis

Stairs are durable building components that maintain the presence of past human activities by displaying signs of wear and tear. Understanding the dynamic trends in historical usage is challenging due to the intricate interplay between mechanical wear, environmental conditions, and incomplete historical documents. To gain a deeper understanding of the usage of historical staircases, this study innovatively adopts a data-driven approach, combining Archard's wear law with convolutional neural networks (CNN), introducing a non-destructive hybrid measurement model that integrates 2D and 3D imaging and mechanical principles, and training CNN with multi angle images to generate wear heat maps for accurate prediction of regional wear; Using Archard's wear law and Poisson distribution to model the relationship between total wear and fine wear, infer historical pedestrian flow intensity, and then analyze directional preferences; Using a continuous force distribution model to divide the staircase area and comparing wear patterns to distinguish between single and double user scenarios to estimate group size; Construct an age estimation model based on wear and tear, taking into account various influencing factors. Overall, the method used in this article not only significantly improves the accuracy of wear analysis, but also provides a scientific and efficient new analytical tool for archaeological research and architectural heritage protection.