EFFECT OF OVERLOADING ON DEVELOPMENT OF DISTRESS IN FLEXIBLE PAVEMENTS

Abstract

The rapid shift from rail to road transport, for movement of passenger and goods traffic, has overstrained the existing large road network in the country. The transporters carry very heavy loads on trucks, majority of which have 2-axle rigid body, without any concern for the damage caused by these to the road. Various factors that affect the damage caused to the pavement include overloading of vehicles, climatic conditions, characteristics of materials used and structural design. The flexible pavement design practice till recently was generally empirical. The limitations of empirical methods of pavement design have become increasingly apparent in recent years with some of the newly constructed highways showing signs of premature failure. The increase in tensile strain at the bottom of bituminous layer results into cracking, whereas accumulation of the compressive strain on top of the subgrade and other layers results into rut formation. Due to overloading the pavement deteriorates faster and its useful life is reduced. The most important types of distress, namely cracking, ravelling, potholes, rutting and roughness, call for prompt maintenance. Cracking and rutting are more important and are characterized by two phases of development - the first one being initiation and the second progression, during which the defects develop in extent and severity. The cracks once initiated propagate upward causing gradual weakening of the pavement structure. The cracking of the bituminous surface is a major mode of distress on Indian roads. Hence, keeping in view the present day traffic and pavement conditions, it was decided to study the effect of overloading of vehicles on development of distress in the form of cracking in flexible pavements. With the main objective of the study in view, the research work was conducted through laboratory, semi-field and field studies, and the analytical work. The various materials used in semi field/ field studies were subjected to laboratory characterization. The testing of bituminous mixes for stiffness modulus (at three temperatures viz. 15, 25 and 35°C) and fatigue characteristics (at 25°C) was done on Nottingham Asphalt Tester (NAT). Under semi-field studies, three sections were constructed in the test hall of Central Road Research Institute (CRRI). These three sections were further divided into 6 subsections, each of 3m x2.5 msize. Out of the six subsections, three were treated as replicates. Instruments were embedded in different layers in all the subsections. The strain gauges were embedded under the bituminous layer and on top of the subgrade, whereas the pressure cells were laid at interfaces of different specifications. Plate load tests were conducted on top surface with a 30 cm diameter plate through reaction loading from a concrete beam. Stresses, strains and displacements were recorded at different load intervals. The performance monitoring of 16 experimental sections, with bituminous concrete surfacing, newly constructed in the field under a research project on four laning works of selected National Highways, was conducted. Two of these sections were embedded with instruments also in different layers. The measurements of stresses and strains were taken, by running a truck at slow speed with rear axle wheel loads of 40.8, 50.1 and 71.4 kN. After opening the sections to regular traffic, periodic performance observations were taken. The periodic performance monitoring included measurements of deflection, distress, traffic characteristics, and subgrade moisture. Acommercially available finite element package has been used for analytical study of the pavements. Anumber of pavement compositions, with different material specifications and properties in the different layers, have been analyzed for various wheel loads. The wheel loads considered are 40.8 (standard wheel loads), 51.0, 71.4 and 81.6 kN (25%, 75% and 100% respectively more than the standard load). The non-linearity on account of the materials has been considered in the analysis. Afinite element model comprising 454 nodes and 135 elements, with axisymmetric loading, was used for analysis of the stresses, strains and displacements in the pavement structure. Analysis has been done both for three layer and five layer structures. The effect of overloading, as compared to standard wheel load, on the various parameters has been evaluated. The data generated from various studies was analyzed. The effect of increase in temperature on the bituminous mixes has been investigated. Through regression analysis of fatigue data, correlations have been developed to determine the amount of tensile strains at failure stage for different axle repetitions. The correlations have been developed separately for different bituminous layers viz., bituminous concrete (BC), semidense bituminous concrete (SDBC) and dense bituminous macadam (DBM). The data generated from analytical and semi-field studies has been analyzed and correlations developed between the applied load and the tensile strain developed at the bottom of bituminous layer. In order to study the effect of overloading, cumulative repetitions in terms of equivalent standard axle load of 81.6 kN (CSALs), that passed on sections at different intervals, were calculated from the traffic and axle load data. The periodic performance data from 16 field sections has been analyzed for development of crack initiation and crack progression models. The surface deflection has been taken as an indicator of pavement strength. Predictive plots have been drawn for different values of surface deflection and traffic repetitions for crack initiation and crack progression of the flexible pavements. A user-friendly computer code "POLAN" has been developed, in Fortran language, to be run on any PC with Window operating system. It enables estimation of pavement life from the results of simple and easy to conduct semi-field / analytical studies and the fatigue tests on the bituminous mixes in the laboratory. With the help of the package and from the given database on pavement surface deflection and traffic characteristics, age in terms of the number of repetitions and also in years, is calculated at which 25 percent surface area would have developed cracks. The limit of 25 percent cracked area is fixed to indicate the failure stage. The effect of overloading on reduction in useful life of flexible pavement can be evaluated from 'POLAN'. IV