Asphalt pavement performance prediction is an important issue for pavement management system. However, it is a difficult problem because asphalt pavement performance are affected by many factors. In this paper, a new method based on fractional gray model is proposed to predict the asphalt pavement performance with a limited data. The proposed method adopts fractional accumulating generating operation (FAGO) to replace traditional accumulating generating operation (AGO), which can be regarded as a weighted AGO emphasizing different contribution of data point for future prediction. An efficient differential evolution algorithm is adopted to select the best order of FAGO. Experimental results show that the proposed method can achieve higher prediction accuracy than conventional gray prediction model. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

中文翻译：

---

使用分数阶 GM(1,1) 模型的路面状况指数预测

沥青路面性能预测是路面管理系统的一个重要问题。然而，这是一个难题，因为沥青路面的性能受多种因素的影响。在本文中，提出了一种基于分数灰色模型的新方法，用于在有限数据下预测沥青路面性能。所提出的方法采用分数累加生成操作（FAGO）来代替传统的累加生成操作（AGO），可以将其视为加权AGO，强调数据点对未来预测的不同贡献。采用一种高效的差分进化算法来选择 FAGO 的最佳阶数。实验结果表明，与传统的灰色预测模型相比，该方法可以实现更高的预测精度。© 2021 日本电气工程师学会。