|dc.description.abstract||Axle load of heavy vehicles is considered as one of the most important loadings exerted on highway pavements. This dissertation aims at investigating the effects of heavy vehicles on the distress and performance of flexible pavements in Taiwan.
Based on field axle load survey data obtained from 32,459 heavy vehicles at the 4 toll stations on Sun Yat-Sen Freeway in Taiwan, a statistical approach was adopted to characterize axle load distribution of heavy vehicles for pavement design and analysis. Procedures for the calculation of full axle load distribution for single and tandem axles are described, and relevant issues on characterizing axle load distribution data were examined. Statistical analyses show that the distribution patterns at the 4 toll stations, for both single and tandem axles, are significantly different. It was found that the cumulative frequency distribution of single and tandem axle load could best be approximated by the Weibull distribution function. Prediction models were developed and validated, using actual field data, for single and tandem axle load distributions on the national freeway.
Multi-layer elastic theory was used in this study to evaluate the effects of heavy vehicle and high tire pressure on the critical strains which are indicative of pavement distress. With the fatigue cracking and rutting failure models, the equivalent axle load factors were determined for flexible pavements in Taiwan. The results show that, as the axle load increases, higher tire pressure will cause marked increase in the equivalent axle load factor, while fatigue failure is the predominant failure mode. If the axle load continues to increase, the failure mode turns to a rutting one and, in such cases, the effect of increases in tire pressure can be ignored.
In the development of pavement distress model, a mechanistic-empirical (M-E) model incorporating a complete list of field and mechanistic parameters was adopted. Using pertinent information available in DataPave 3.0, regression analyses were conducted to determine the parameters that are significant for the distress model. Consequently, two performance-based pavement evaluation models adaptable to local material and traffic characteristics are established, including the rut prediction model and fatigue life model. Also, a pavement performance life model, which is of great convenience for pavement management purposes in highway agencies, is proposed in terms of international roughness index (IRI). Finally, the performance model was validated using limited field test data available from Sun Yat-Sun Freeway, such that the procedures for further modification of the model for future applications in Taiwan have been advised.||en_US|