| Summary: | 碩士 === 淡江大學 === 土木工程學系碩士班 === 99 === Pavement performance data is a very common example of multilevel data. While analyzing this type of data using conventional regression techniques, the normality assumptions with random errors and constant variance were often violated. Because of its hierarchical data structure, multilevel data are often analyzed using Linear Mixed-Effects (LME) models. The exploratory analysis, statistical modeling, and the examination of model-fit of LME models are more complicated than those of standard multiple regressions. A preliminary LME model for PSI prediction was developed by Huang (2010) using the original AASHO road test flexible pavement data.
This is a continuous study to explore and validate the applicability of the aforementioned preliminary LME model particularly on the potential use of equivalent axle load factors (EALF) or load equivalency factors (LEF). Necessary steps have been made to enhance the existing LME model. In addition, projection pursuit regression, local regression and nonlinear regression techniques were also adopted in an attempt to develop modified flexible pavement design equations for single- and tandem- axle loads separately.
Various load equivalency factors have been derived using different predictive models and compared to the existing LEFs of the AASHTO guides. Even though reasonable results have been obtained, the newly derived LEFs representing quite a departure from the well-known fourth-power rule should be cautioned and further investigated.
|
|---|
