- 期刊
Development of Rut Prediction Models from an Instrumented In-Service Test Section on Interstate-35
並列摘要
This paper presents development of two different field rut prediction models based on data collected from an instrumented pavement section on I-35 in McClain County, Oklahoma. Two rut prediction models, vertical strain-based (VSB) and shear strain-based (SSB), were developed utilizing four years of pavement and environmental data and from approximately 18.7-million accumulated axles. The VSB model considers vertical strain on the top of the aggregate base layer, while the SSB model was based on the shear strain in the Hot Mix Asphalt (HMA) layer. Falling Weight Deflectometer (FWD) tests were conducted over a wide range of temperature to establish modulus and temperature relationship. A pavement analysis software, WinJULEA, was used to develop correlations between temperature and vertical and shear strains for single and tandem axles. In addition, field rut measurements were conducted periodically using a straight edge-rut gauge combination and a Face Dipstick^®. A systematic methodology to develop the rut prediction models is presented in the paper. The correlation coefficient (R^2) for the VSB and the SSB models were found to be 0.78 and 0.72, respectively. Statistical analyses showed that both models predicted rut with a similar level of accuracy. The results from this study are expected to be useful in predicting rut of state highway pavements under similar traffic and environmental conditions. In addition, data collected from this study may be used for local calibration of rut prediction model available in the Mechanistic-Empirical Pavement Design Guide (MEPDG).