Scholarlink

<p>A scientific approach is essential for evaluating pavement surface conditions at the network level. The pavement condition index (PCI) is widely used to assess surface conditions and determine appropriate treatments. This study examines three national highways using a network survey vehicle to collect distress data. The first two corridors were used for evaluation and comparison, while the third corridor validated the predicted PCI values. Multiple linear regression initially modeled the relationship between PCI and distress variables but showed poor predictive accuracy. Therefore, K-nearest neighbors, artificial neural network, and support vector machine models were developed, providing better results. A methodology for prioritizing pavement sections was introduced, and the pavement sections were based on PCI, International Roughness Index (IRI), and rut values through Combined Index Rankings.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development of PCI Prediction Models for Asphalt Pavements

  • 期刊论文

  • 发表日期：

    2025年9月9日

查看全文
  • Soma Prashanth Kumar,
  • R. Srinivasa Kumar,
  • Hamid Noori

摘要

A scientific approach is essential for evaluating pavement surface conditions at the network level. The pavement condition index (PCI) is widely used to assess surface conditions and determine appropriate treatments. This study examines three national highways using a network survey vehicle to collect distress data. The first two corridors were used for evaluation and comparison, while the third corridor validated the predicted PCI values. Multiple linear regression initially modeled the relationship between PCI and distress variables but showed poor predictive accuracy. Therefore, K-nearest neighbors, artificial neural network, and support vector machine models were developed, providing better results. A methodology for prioritizing pavement sections was introduced, and the pavement sections were based on PCI, International Roughness Index (IRI), and rut values through Combined Index Rankings.